init

attentions.py

@@ -0,0 +1,303 @@
+import copy
+import math
+import numpy as np
+import torch
+from torch import nn
+from torch.nn import functional as F
+
+import commons
+import modules
+from modules import LayerNorm
+   
+
+class Encoder(nn.Module):
+  def __init__(self, hidden_channels, filter_channels, n_heads, n_layers, kernel_size=1, p_dropout=0., window_size=4, **kwargs):
+    super().__init__()
+    self.hidden_channels = hidden_channels
+    self.filter_channels = filter_channels
+    self.n_heads = n_heads
+    self.n_layers = n_layers
+    self.kernel_size = kernel_size
+    self.p_dropout = p_dropout
+    self.window_size = window_size
+
+    self.drop = nn.Dropout(p_dropout)
+    self.attn_layers = nn.ModuleList()
+    self.norm_layers_1 = nn.ModuleList()
+    self.ffn_layers = nn.ModuleList()
+    self.norm_layers_2 = nn.ModuleList()
+    for i in range(self.n_layers):
+      self.attn_layers.append(MultiHeadAttention(hidden_channels, hidden_channels, n_heads, p_dropout=p_dropout, window_size=window_size))
+      self.norm_layers_1.append(LayerNorm(hidden_channels))
+      self.ffn_layers.append(FFN(hidden_channels, hidden_channels, filter_channels, kernel_size, p_dropout=p_dropout))
+      self.norm_layers_2.append(LayerNorm(hidden_channels))
+
+  def forward(self, x, x_mask):
+    attn_mask = x_mask.unsqueeze(2) * x_mask.unsqueeze(-1)
+    x = x * x_mask
+    for i in range(self.n_layers):
+      y = self.attn_layers[i](x, x, attn_mask)
+      y = self.drop(y)
+      x = self.norm_layers_1[i](x + y)
+
+      y = self.ffn_layers[i](x, x_mask)
+      y = self.drop(y)
+      x = self.norm_layers_2[i](x + y)
+    x = x * x_mask
+    return x
+
+
+class Decoder(nn.Module):
+  def __init__(self, hidden_channels, filter_channels, n_heads, n_layers, kernel_size=1, p_dropout=0., proximal_bias=False, proximal_init=True, **kwargs):
+    super().__init__()
+    self.hidden_channels = hidden_channels
+    self.filter_channels = filter_channels
+    self.n_heads = n_heads
+    self.n_layers = n_layers
+    self.kernel_size = kernel_size
+    self.p_dropout = p_dropout
+    self.proximal_bias = proximal_bias
+    self.proximal_init = proximal_init
+
+    self.drop = nn.Dropout(p_dropout)
+    self.self_attn_layers = nn.ModuleList()
+    self.norm_layers_0 = nn.ModuleList()
+    self.encdec_attn_layers = nn.ModuleList()
+    self.norm_layers_1 = nn.ModuleList()
+    self.ffn_layers = nn.ModuleList()
+    self.norm_layers_2 = nn.ModuleList()
+    for i in range(self.n_layers):
+      self.self_attn_layers.append(MultiHeadAttention(hidden_channels, hidden_channels, n_heads, p_dropout=p_dropout, proximal_bias=proximal_bias, proximal_init=proximal_init))
+      self.norm_layers_0.append(LayerNorm(hidden_channels))
+      self.encdec_attn_layers.append(MultiHeadAttention(hidden_channels, hidden_channels, n_heads, p_dropout=p_dropout))
+      self.norm_layers_1.append(LayerNorm(hidden_channels))
+      self.ffn_layers.append(FFN(hidden_channels, hidden_channels, filter_channels, kernel_size, p_dropout=p_dropout, causal=True))
+      self.norm_layers_2.append(LayerNorm(hidden_channels))
+
+  def forward(self, x, x_mask, h, h_mask):
+    """
+    x: decoder input
+    h: encoder output
+    """
+    self_attn_mask = commons.subsequent_mask(x_mask.size(2)).to(device=x.device, dtype=x.dtype)
+    encdec_attn_mask = h_mask.unsqueeze(2) * x_mask.unsqueeze(-1)
+    x = x * x_mask
+    for i in range(self.n_layers):
+      y = self.self_attn_layers[i](x, x, self_attn_mask)
+      y = self.drop(y)
+      x = self.norm_layers_0[i](x + y)
+
+      y = self.encdec_attn_layers[i](x, h, encdec_attn_mask)
+      y = self.drop(y)
+      x = self.norm_layers_1[i](x + y)
+      
+      y = self.ffn_layers[i](x, x_mask)
+      y = self.drop(y)
+      x = self.norm_layers_2[i](x + y)
+    x = x * x_mask
+    return x
+
+
+class MultiHeadAttention(nn.Module):
+  def __init__(self, channels, out_channels, n_heads, p_dropout=0., window_size=None, heads_share=True, block_length=None, proximal_bias=False, proximal_init=False):
+    super().__init__()
+    assert channels % n_heads == 0
+
+    self.channels = channels
+    self.out_channels = out_channels
+    self.n_heads = n_heads
+    self.p_dropout = p_dropout
+    self.window_size = window_size
+    self.heads_share = heads_share
+    self.block_length = block_length
+    self.proximal_bias = proximal_bias
+    self.proximal_init = proximal_init
+    self.attn = None
+
+    self.k_channels = channels // n_heads
+    self.conv_q = nn.Conv1d(channels, channels, 1)
+    self.conv_k = nn.Conv1d(channels, channels, 1)
+    self.conv_v = nn.Conv1d(channels, channels, 1)
+    self.conv_o = nn.Conv1d(channels, out_channels, 1)
+    self.drop = nn.Dropout(p_dropout)
+
+    if window_size is not None:
+      n_heads_rel = 1 if heads_share else n_heads
+      rel_stddev = self.k_channels**-0.5
+      self.emb_rel_k = nn.Parameter(torch.randn(n_heads_rel, window_size * 2 + 1, self.k_channels) * rel_stddev)
+      self.emb_rel_v = nn.Parameter(torch.randn(n_heads_rel, window_size * 2 + 1, self.k_channels) * rel_stddev)
+
+    nn.init.xavier_uniform_(self.conv_q.weight)
+    nn.init.xavier_uniform_(self.conv_k.weight)
+    nn.init.xavier_uniform_(self.conv_v.weight)
+    if proximal_init:
+      with torch.no_grad():
+        self.conv_k.weight.copy_(self.conv_q.weight)
+        self.conv_k.bias.copy_(self.conv_q.bias)
+      
+  def forward(self, x, c, attn_mask=None):
+    q = self.conv_q(x)
+    k = self.conv_k(c)
+    v = self.conv_v(c)
+    
+    x, self.attn = self.attention(q, k, v, mask=attn_mask)
+
+    x = self.conv_o(x)
+    return x
+
+  def attention(self, query, key, value, mask=None):
+    # reshape [b, d, t] -> [b, n_h, t, d_k]
+    b, d, t_s, t_t = (*key.size(), query.size(2))
+    query = query.view(b, self.n_heads, self.k_channels, t_t).transpose(2, 3)
+    key = key.view(b, self.n_heads, self.k_channels, t_s).transpose(2, 3)
+    value = value.view(b, self.n_heads, self.k_channels, t_s).transpose(2, 3)
+
+    scores = torch.matmul(query / math.sqrt(self.k_channels), key.transpose(-2, -1))
+    if self.window_size is not None:
+      assert t_s == t_t, "Relative attention is only available for self-attention."
+      key_relative_embeddings = self._get_relative_embeddings(self.emb_rel_k, t_s)
+      rel_logits = self._matmul_with_relative_keys(query /math.sqrt(self.k_channels), key_relative_embeddings)
+      scores_local = self._relative_position_to_absolute_position(rel_logits)
+      scores = scores + scores_local
+    if self.proximal_bias:
+      assert t_s == t_t, "Proximal bias is only available for self-attention."
+      scores = scores + self._attention_bias_proximal(t_s).to(device=scores.device, dtype=scores.dtype)
+    if mask is not None:
+      scores = scores.masked_fill(mask == 0, -1e4)
+      if self.block_length is not None:
+        assert t_s == t_t, "Local attention is only available for self-attention."
+        block_mask = torch.ones_like(scores).triu(-self.block_length).tril(self.block_length)
+        scores = scores.masked_fill(block_mask == 0, -1e4)
+    p_attn = F.softmax(scores, dim=-1) # [b, n_h, t_t, t_s]
+    p_attn = self.drop(p_attn)
+    output = torch.matmul(p_attn, value)
+    if self.window_size is not None:
+      relative_weights = self._absolute_position_to_relative_position(p_attn)
+      value_relative_embeddings = self._get_relative_embeddings(self.emb_rel_v, t_s)
+      output = output + self._matmul_with_relative_values(relative_weights, value_relative_embeddings)
+    output = output.transpose(2, 3).contiguous().view(b, d, t_t) # [b, n_h, t_t, d_k] -> [b, d, t_t]
+    return output, p_attn
+
+  def _matmul_with_relative_values(self, x, y):
+    """
+    x: [b, h, l, m]
+    y: [h or 1, m, d]
+    ret: [b, h, l, d]
+    """
+    ret = torch.matmul(x, y.unsqueeze(0))
+    return ret
+
+  def _matmul_with_relative_keys(self, x, y):
+    """
+    x: [b, h, l, d]
+    y: [h or 1, m, d]
+    ret: [b, h, l, m]
+    """
+    ret = torch.matmul(x, y.unsqueeze(0).transpose(-2, -1))
+    return ret
+
+  def _get_relative_embeddings(self, relative_embeddings, length):
+    max_relative_position = 2 * self.window_size + 1
+    # Pad first before slice to avoid using cond ops.
+    pad_length = max(length - (self.window_size + 1), 0)
+    slice_start_position = max((self.window_size + 1) - length, 0)
+    slice_end_position = slice_start_position + 2 * length - 1
+    if pad_length > 0:
+      padded_relative_embeddings = F.pad(
+          relative_embeddings,
+          commons.convert_pad_shape([[0, 0], [pad_length, pad_length], [0, 0]]))
+    else:
+      padded_relative_embeddings = relative_embeddings
+    used_relative_embeddings = padded_relative_embeddings[:,slice_start_position:slice_end_position]
+    return used_relative_embeddings
+
+  def _relative_position_to_absolute_position(self, x):
+    """
+    x: [b, h, l, 2*l-1]
+    ret: [b, h, l, l]
+    """
+    batch, heads, length, _ = x.size()
+    # Concat columns of pad to shift from relative to absolute indexing.
+    x = F.pad(x, commons.convert_pad_shape([[0,0],[0,0],[0,0],[0,1]]))
+
+    # Concat extra elements so to add up to shape (len+1, 2*len-1).
+    x_flat = x.view([batch, heads, length * 2 * length])
+    x_flat = F.pad(x_flat, commons.convert_pad_shape([[0,0],[0,0],[0,length-1]]))
+
+    # Reshape and slice out the padded elements.
+    x_final = x_flat.view([batch, heads, length+1, 2*length-1])[:, :, :length, length-1:]
+    return x_final
+
+  def _absolute_position_to_relative_position(self, x):
+    """
+    x: [b, h, l, l]
+    ret: [b, h, l, 2*l-1]
+    """
+    batch, heads, length, _ = x.size()
+    # padd along column
+    x = F.pad(x, commons.convert_pad_shape([[0, 0], [0, 0], [0, 0], [0, length-1]]))
+    x_flat = x.view([batch, heads, length**2 + length*(length -1)])
+    # add 0's in the beginning that will skew the elements after reshape
+    x_flat = F.pad(x_flat, commons.convert_pad_shape([[0, 0], [0, 0], [length, 0]]))
+    x_final = x_flat.view([batch, heads, length, 2*length])[:,:,:,1:]
+    return x_final
+
+  def _attention_bias_proximal(self, length):
+    """Bias for self-attention to encourage attention to close positions.
+    Args:
+      length: an integer scalar.
+    Returns:
+      a Tensor with shape [1, 1, length, length]
+    """
+    r = torch.arange(length, dtype=torch.float32)
+    diff = torch.unsqueeze(r, 0) - torch.unsqueeze(r, 1)
+    return torch.unsqueeze(torch.unsqueeze(-torch.log1p(torch.abs(diff)), 0), 0)
+
+
+class FFN(nn.Module):
+  def __init__(self, in_channels, out_channels, filter_channels, kernel_size, p_dropout=0., activation=None, causal=False):
+    super().__init__()
+    self.in_channels = in_channels
+    self.out_channels = out_channels
+    self.filter_channels = filter_channels
+    self.kernel_size = kernel_size
+    self.p_dropout = p_dropout
+    self.activation = activation
+    self.causal = causal
+
+    if causal:
+      self.padding = self._causal_padding
+    else:
+      self.padding = self._same_padding
+
+    self.conv_1 = nn.Conv1d(in_channels, filter_channels, kernel_size)
+    self.conv_2 = nn.Conv1d(filter_channels, out_channels, kernel_size)
+    self.drop = nn.Dropout(p_dropout)
+
+  def forward(self, x, x_mask):
+    x = self.conv_1(self.padding(x * x_mask))
+    if self.activation == "gelu":
+      x = x * torch.sigmoid(1.702 * x)
+    else:
+      x = torch.relu(x)
+    x = self.drop(x)
+    x = self.conv_2(self.padding(x * x_mask))
+    return x * x_mask
+  
+  def _causal_padding(self, x):
+    if self.kernel_size == 1:
+      return x
+    pad_l = self.kernel_size - 1
+    pad_r = 0
+    padding = [[0, 0], [0, 0], [pad_l, pad_r]]
+    x = F.pad(x, commons.convert_pad_shape(padding))
+    return x
+
+  def _same_padding(self, x):
+    if self.kernel_size == 1:
+      return x
+    pad_l = (self.kernel_size - 1) // 2
+    pad_r = self.kernel_size // 2
+    padding = [[0, 0], [0, 0], [pad_l, pad_r]]
+    x = F.pad(x, commons.convert_pad_shape(padding))
+    return x

commons.py

@@ -0,0 +1,161 @@
+import math
+import numpy as np
+import torch
+from torch import nn
+from torch.nn import functional as F
+
+
+def init_weights(m, mean=0.0, std=0.01):
+  classname = m.__class__.__name__
+  if classname.find("Conv") != -1:
+    m.weight.data.normal_(mean, std)
+
+
+def get_padding(kernel_size, dilation=1):
+  return int((kernel_size*dilation - dilation)/2)
+
+
+def convert_pad_shape(pad_shape):
+  l = pad_shape[::-1]
+  pad_shape = [item for sublist in l for item in sublist]
+  return pad_shape
+
+
+def intersperse(lst, item):
+  result = [item] * (len(lst) * 2 + 1)
+  result[1::2] = lst
+  return result
+
+
+def kl_divergence(m_p, logs_p, m_q, logs_q):
+  """KL(P||Q)"""
+  kl = (logs_q - logs_p) - 0.5
+  kl += 0.5 * (torch.exp(2. * logs_p) + ((m_p - m_q)**2)) * torch.exp(-2. * logs_q)
+  return kl
+
+
+def rand_gumbel(shape):
+  """Sample from the Gumbel distribution, protect from overflows."""
+  uniform_samples = torch.rand(shape) * 0.99998 + 0.00001
+  return -torch.log(-torch.log(uniform_samples))
+
+
+def rand_gumbel_like(x):
+  g = rand_gumbel(x.size()).to(dtype=x.dtype, device=x.device)
+  return g
+
+
+def slice_segments(x, ids_str, segment_size=4):
+  ret = torch.zeros_like(x[:, :, :segment_size])
+  for i in range(x.size(0)):
+    idx_str = ids_str[i]
+    idx_end = idx_str + segment_size
+    ret[i] = x[i, :, idx_str:idx_end]
+  return ret
+
+
+def rand_slice_segments(x, x_lengths=None, segment_size=4):
+  b, d, t = x.size()
+  if x_lengths is None:
+    x_lengths = t
+  ids_str_max = x_lengths - segment_size + 1
+  ids_str = (torch.rand([b]).to(device=x.device) * ids_str_max).to(dtype=torch.long)
+  ret = slice_segments(x, ids_str, segment_size)
+  return ret, ids_str
+
+
+def get_timing_signal_1d(
+    length, channels, min_timescale=1.0, max_timescale=1.0e4):
+  position = torch.arange(length, dtype=torch.float)
+  num_timescales = channels // 2
+  log_timescale_increment = (
+      math.log(float(max_timescale) / float(min_timescale)) /
+      (num_timescales - 1))
+  inv_timescales = min_timescale * torch.exp(
+      torch.arange(num_timescales, dtype=torch.float) * -log_timescale_increment)
+  scaled_time = position.unsqueeze(0) * inv_timescales.unsqueeze(1)
+  signal = torch.cat([torch.sin(scaled_time), torch.cos(scaled_time)], 0)
+  signal = F.pad(signal, [0, 0, 0, channels % 2])
+  signal = signal.view(1, channels, length)
+  return signal
+
+
+def add_timing_signal_1d(x, min_timescale=1.0, max_timescale=1.0e4):
+  b, channels, length = x.size()
+  signal = get_timing_signal_1d(length, channels, min_timescale, max_timescale)
+  return x + signal.to(dtype=x.dtype, device=x.device)
+
+
+def cat_timing_signal_1d(x, min_timescale=1.0, max_timescale=1.0e4, axis=1):
+  b, channels, length = x.size()
+  signal = get_timing_signal_1d(length, channels, min_timescale, max_timescale)
+  return torch.cat([x, signal.to(dtype=x.dtype, device=x.device)], axis)
+
+
+def subsequent_mask(length):
+  mask = torch.tril(torch.ones(length, length)).unsqueeze(0).unsqueeze(0)
+  return mask
+
+
+@torch.jit.script
+def fused_add_tanh_sigmoid_multiply(input_a, input_b, n_channels):
+  n_channels_int = n_channels[0]
+  in_act = input_a + input_b
+  t_act = torch.tanh(in_act[:, :n_channels_int, :])
+  s_act = torch.sigmoid(in_act[:, n_channels_int:, :])
+  acts = t_act * s_act
+  return acts
+
+
+def convert_pad_shape(pad_shape):
+  l = pad_shape[::-1]
+  pad_shape = [item for sublist in l for item in sublist]
+  return pad_shape
+
+
+def shift_1d(x):
+  x = F.pad(x, convert_pad_shape([[0, 0], [0, 0], [1, 0]]))[:, :, :-1]
+  return x
+
+
+def sequence_mask(length, max_length=None):
+  if max_length is None:
+    max_length = length.max()
+  x = torch.arange(max_length, dtype=length.dtype, device=length.device)
+  return x.unsqueeze(0) < length.unsqueeze(1)
+
+
+def generate_path(duration, mask):
+  """
+  duration: [b, 1, t_x]
+  mask: [b, 1, t_y, t_x]
+  """
+  device = duration.device
+  
+  b, _, t_y, t_x = mask.shape
+  cum_duration = torch.cumsum(duration, -1)
+  
+  cum_duration_flat = cum_duration.view(b * t_x)
+  path = sequence_mask(cum_duration_flat, t_y).to(mask.dtype)
+  path = path.view(b, t_x, t_y)
+  path = path - F.pad(path, convert_pad_shape([[0, 0], [1, 0], [0, 0]]))[:, :-1]
+  path = path.unsqueeze(1).transpose(2,3) * mask
+  return path
+
+
+def clip_grad_value_(parameters, clip_value, norm_type=2):
+  if isinstance(parameters, torch.Tensor):
+    parameters = [parameters]
+  parameters = list(filter(lambda p: p.grad is not None, parameters))
+  norm_type = float(norm_type)
+  if clip_value is not None:
+    clip_value = float(clip_value)
+
+  total_norm = 0
+  for p in parameters:
+    param_norm = p.grad.data.norm(norm_type)
+    total_norm += param_norm.item() ** norm_type
+    if clip_value is not None:
+      p.grad.data.clamp_(min=-clip_value, max=clip_value)
+  total_norm = total_norm ** (1. / norm_type)
+  return total_norm

configs/config.json

@@ -0,0 +1,88 @@
+{
+  "train": {
+    "log_interval": 200,
+    "eval_interval": 1000,
+    "seed": 1234,
+    "epochs": 10000,
+    "learning_rate": 0.0001,
+    "betas": [
+      0.8,
+      0.99
+    ],
+    "eps": 1e-09,
+    "batch_size": 32,
+    "fp16_run": true,
+    "lr_decay": 0.999875,
+    "segment_size": 8192,
+    "init_lr_ratio": 1,
+    "warmup_epochs": 0,
+    "c_mel": 45,
+    "c_kl": 1.0
+  },
+  "data": {
+    "training_files": "filelists/train.list",
+    "validation_files": "filelists/val.list",
+    "max_wav_value": 32768.0,
+    "sampling_rate": 22050,
+    "filter_length": 1024,
+    "hop_length": 256,
+    "win_length": 1024,
+    "n_mel_channels": 80,
+    "mel_fmin": 0.0,
+    "mel_fmax": null,
+    "add_blank": true,
+    "n_speakers": 300,
+    "cleaned_text": true,
+    "spk2id": {
+      "0001_Angry": 0
+    }
+  },
+  "model": {
+    "inter_channels": 192,
+    "hidden_channels": 192,
+    "filter_channels": 768,
+    "n_heads": 2,
+    "n_layers": 6,
+    "kernel_size": 3,
+    "p_dropout": 0.1,
+    "resblock": "1",
+    "resblock_kernel_sizes": [
+      3,
+      7,
+      11
+    ],
+    "resblock_dilation_sizes": [
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ],
+      [
+        1,
+        3,
+        5
+      ]
+    ],
+    "upsample_rates": [
+      8,
+      8,
+      2,
+      2
+    ],
+    "upsample_initial_channel": 512,
+    "upsample_kernel_sizes": [
+      16,
+      16,
+      4,
+      4
+    ],
+    "n_layers_q": 3,
+    "use_spectral_norm": false,
+    "gin_channels": 256
+  }
+}

data_utils.py

@@ -0,0 +1,282 @@
+import time
+import os
+import random
+import numpy as np
+import torch
+import torch.utils.data
+
+import commons 
+from mel_processing import spectrogram_torch
+from utils import load_wav_to_torch, load_filepaths_and_text
+from text import cleaned_text_to_sequence, cleaned_text_to_sequence_bert, get_bert
+
+"""Multi speaker version"""
+class TextAudioSpeakerLoader(torch.utils.data.Dataset):
+    """
+        1) loads audio, speaker_id, text pairs
+        2) normalizes text and converts them to sequences of integers
+        3) computes spectrograms from audio files.
+    """
+    def __init__(self, audiopaths_sid_text, hparams):
+        self.audiopaths_sid_text = load_filepaths_and_text(audiopaths_sid_text)
+        self.max_wav_value = hparams.max_wav_value
+        self.sampling_rate = hparams.sampling_rate
+        self.filter_length  = hparams.filter_length
+        self.hop_length     = hparams.hop_length
+        self.win_length     = hparams.win_length
+        self.sampling_rate  = hparams.sampling_rate
+        self.spk_map = hparams.spk2id
+
+        self.cleaned_text = getattr(hparams, "cleaned_text", False)
+
+        self.add_blank = hparams.add_blank
+        self.min_text_len = getattr(hparams, "min_text_len", 1)
+        self.max_text_len = getattr(hparams, "max_text_len", 300)
+
+        random.seed(1234)
+        random.shuffle(self.audiopaths_sid_text)
+        self._filter()
+
+    def _filter(self):
+        """
+        Filter text & store spec lengths
+        """
+        # Store spectrogram lengths for Bucketing
+        # wav_length ~= file_size / (wav_channels * Bytes per dim) = file_size / (1 * 2)
+        # spec_length = wav_length // hop_length
+
+        audiopaths_sid_text_new = []
+        lengths = []
+        skipped = 0
+        for _id, spk, language, text, phones, tone, word2ph in self.audiopaths_sid_text:
+            audiopath =  f'dataset/{spk}/{_id}.wav'
+            if self.min_text_len <= len(phones) and len(phones) <= self.max_text_len:
+                phones = phones.split(" ")
+                tone = [int(i) for i in tone.split(" ")]
+                word2ph = [int(i) for i in word2ph.split(" ")]
+                audiopaths_sid_text_new.append([audiopath, spk, language,text, phones, tone, word2ph])
+                lengths.append(os.path.getsize(audiopath) // (2 * self.hop_length))
+            else:
+                skipped += 1
+        print("skipped: ", skipped, ", total: ", len(self.audiopaths_sid_text))
+        self.audiopaths_sid_text = audiopaths_sid_text_new
+        self.lengths = lengths
+
+    def get_audio_text_speaker_pair(self, audiopath_sid_text):
+        # separate filename, speaker_id and text
+        audiopath, sid, language, text, phones, tone, word2ph = audiopath_sid_text
+
+        bert, phones, tone, language = self.get_text(text, word2ph, phones, tone, language)
+
+        spec, wav = self.get_audio(audiopath)
+        sid = torch.LongTensor([int(self.spk_map[sid])])
+        return (phones, spec, wav, sid, tone, language, bert)
+
+    def get_audio(self, filename):
+        audio, sampling_rate = load_wav_to_torch(filename)
+        if sampling_rate != self.sampling_rate:
+            raise ValueError("{} {} SR doesn't match target {} SR".format(
+                sampling_rate, self.sampling_rate))
+        audio_norm = audio / self.max_wav_value
+        audio_norm = audio_norm.unsqueeze(0)
+        spec_filename = filename.replace(".wav", ".spec.pt")
+        if os.path.exists(spec_filename):
+            spec = torch.load(spec_filename)
+        else:
+            spec = spectrogram_torch(audio_norm, self.filter_length,
+                self.sampling_rate, self.hop_length, self.win_length,
+                center=False)
+            spec = torch.squeeze(spec, 0)
+            torch.save(spec, spec_filename)
+        return spec, audio_norm
+
+    def get_text(self,text, word2ph,phone, tone, language_str):
+        phone, tone, language = cleaned_text_to_sequence(phone, tone, language_str)
+        if self.add_blank:
+            phone = commons.intersperse(phone, 0)
+            tone = commons.intersperse(tone, 0)
+            language = commons.intersperse(language, 0)
+            for i in range(1,len(word2ph)):
+                word2ph[i] += 1
+        bert = get_bert(text, word2ph, language_str)
+        phone = torch.LongTensor(phone)
+        tone = torch.LongTensor(tone)
+        language = torch.LongTensor(language)
+        return bert, phone, tone, language
+
+    def get_sid(self, sid):
+        sid = torch.LongTensor([int(sid)])
+        return sid
+
+    def __getitem__(self, index):
+        return self.get_audio_text_speaker_pair(self.audiopaths_sid_text[index])
+
+    def __len__(self):
+        return len(self.audiopaths_sid_text)
+
+
+class TextAudioSpeakerCollate():
+    """ Zero-pads model inputs and targets
+    """
+    def __init__(self, return_ids=False):
+        self.return_ids = return_ids
+
+    def __call__(self, batch):
+        """Collate's training batch from normalized text, audio and speaker identities
+        PARAMS
+        ------
+        batch: [text_normalized, spec_normalized, wav_normalized, sid]
+        """
+        # Right zero-pad all one-hot text sequences to max input length
+        _, ids_sorted_decreasing = torch.sort(
+            torch.LongTensor([x[1].size(1) for x in batch]),
+            dim=0, descending=True)
+
+        max_text_len = max([len(x[0]) for x in batch])
+        max_spec_len = max([x[1].size(1) for x in batch])
+        max_wav_len = max([x[2].size(1) for x in batch])
+
+        text_lengths = torch.LongTensor(len(batch))
+        spec_lengths = torch.LongTensor(len(batch))
+        wav_lengths = torch.LongTensor(len(batch))
+        sid = torch.LongTensor(len(batch))
+
+        text_padded = torch.LongTensor(len(batch), max_text_len)
+        tone_padded = torch.LongTensor(len(batch), max_text_len)
+        language_padded = torch.LongTensor(len(batch), max_text_len)
+        bert_padded = torch.FloatTensor(len(batch), 1024, max_text_len)
+
+        spec_padded = torch.FloatTensor(len(batch), batch[0][1].size(0), max_spec_len)
+        wav_padded = torch.FloatTensor(len(batch), 1, max_wav_len)
+        text_padded.zero_()
+        tone_padded.zero_()
+        language_padded.zero_()
+        spec_padded.zero_()
+        wav_padded.zero_()
+        bert_padded.zero_()
+        for i in range(len(ids_sorted_decreasing)):
+            row = batch[ids_sorted_decreasing[i]]
+
+            text = row[0]
+            text_padded[i, :text.size(0)] = text
+            text_lengths[i] = text.size(0)
+
+            spec = row[1]
+            spec_padded[i, :, :spec.size(1)] = spec
+            spec_lengths[i] = spec.size(1)
+
+            wav = row[2]
+            wav_padded[i, :, :wav.size(1)] = wav
+            wav_lengths[i] = wav.size(1)
+
+            sid[i] = row[3]
+
+            tone = row[4]
+            tone_padded[i, :tone.size(0)] = tone
+
+            language = row[5]
+            language_padded[i, :language.size(0)] = language
+
+            bert = row[6]
+            bert_padded[i, :, :bert.size(1)] = bert
+
+        return text_padded, text_lengths, spec_padded, spec_lengths, wav_padded, wav_lengths, sid, tone_padded, language_padded, bert_padded
+
+
+class DistributedBucketSampler(torch.utils.data.distributed.DistributedSampler):
+    """
+    Maintain similar input lengths in a batch.
+    Length groups are specified by boundaries.
+    Ex) boundaries = [b1, b2, b3] -> any batch is included either {x | b1 < length(x) <=b2} or {x | b2 < length(x) <= b3}.
+  
+    It removes samples which are not included in the boundaries.
+    Ex) boundaries = [b1, b2, b3] -> any x s.t. length(x) <= b1 or length(x) > b3 are discarded.
+    """
+    def __init__(self, dataset, batch_size, boundaries, num_replicas=None, rank=None, shuffle=True):
+        super().__init__(dataset, num_replicas=num_replicas, rank=rank, shuffle=shuffle)
+        self.lengths = dataset.lengths
+        self.batch_size = batch_size
+        self.boundaries = boundaries
+  
+        self.buckets, self.num_samples_per_bucket = self._create_buckets()
+        self.total_size = sum(self.num_samples_per_bucket)
+        self.num_samples = self.total_size // self.num_replicas
+  
+    def _create_buckets(self):
+        buckets = [[] for _ in range(len(self.boundaries) - 1)]
+        for i in range(len(self.lengths)):
+            length = self.lengths[i]
+            idx_bucket = self._bisect(length)
+            if idx_bucket != -1:
+                buckets[idx_bucket].append(i)
+  
+        for i in range(len(buckets) - 1, 0, -1):
+            if len(buckets[i]) == 0:
+                buckets.pop(i)
+                self.boundaries.pop(i+1)
+  
+        num_samples_per_bucket = []
+        for i in range(len(buckets)):
+            len_bucket = len(buckets[i])
+            total_batch_size = self.num_replicas * self.batch_size
+            rem = (total_batch_size - (len_bucket % total_batch_size)) % total_batch_size
+            num_samples_per_bucket.append(len_bucket + rem)
+        return buckets, num_samples_per_bucket
+  
+    def __iter__(self):
+      # deterministically shuffle based on epoch
+      g = torch.Generator()
+      g.manual_seed(self.epoch)
+  
+      indices = []
+      if self.shuffle:
+          for bucket in self.buckets:
+              indices.append(torch.randperm(len(bucket), generator=g).tolist())
+      else:
+          for bucket in self.buckets:
+              indices.append(list(range(len(bucket))))
+  
+      batches = []
+      for i in range(len(self.buckets)):
+          bucket = self.buckets[i]
+          len_bucket = len(bucket)
+          ids_bucket = indices[i]
+          num_samples_bucket = self.num_samples_per_bucket[i]
+  
+          # add extra samples to make it evenly divisible
+          rem = num_samples_bucket - len_bucket
+          ids_bucket = ids_bucket + ids_bucket * (rem // len_bucket) + ids_bucket[:(rem % len_bucket)]
+  
+          # subsample
+          ids_bucket = ids_bucket[self.rank::self.num_replicas]
+  
+          # batching
+          for j in range(len(ids_bucket) // self.batch_size):
+              batch = [bucket[idx] for idx in ids_bucket[j*self.batch_size:(j+1)*self.batch_size]]
+              batches.append(batch)
+  
+      if self.shuffle:
+          batch_ids = torch.randperm(len(batches), generator=g).tolist()
+          batches = [batches[i] for i in batch_ids]
+      self.batches = batches
+  
+      assert len(self.batches) * self.batch_size == self.num_samples
+      return iter(self.batches)
+  
+    def _bisect(self, x, lo=0, hi=None):
+      if hi is None:
+          hi = len(self.boundaries) - 1
+  
+      if hi > lo:
+          mid = (hi + lo) // 2
+          if self.boundaries[mid] < x and x <= self.boundaries[mid+1]:
+              return mid
+          elif x <= self.boundaries[mid]:
+              return self._bisect(x, lo, mid)
+          else:
+              return self._bisect(x, mid + 1, hi)
+      else:
+          return -1
+
+    def __len__(self):
+        return self.num_samples // self.batch_size

filelists/esd.list

@@ -0,0 +1,243 @@
+0001_000352|0001_Angry|ZH|英国的哲学家曾经说过“
+0001_000353|0001_Angry|ZH|我老家在北京，哇塞！太精彩了。
+0001_000351|0001_Angry|ZH|打远一看，它们的确很是美丽，
+0001_000354|0001_Angry|ZH|不管怎么说主队好象是志在夺魁。
+0001_000368|0001_Angry|ZH|我们意见不和，咱们去那儿玩吧。
+0001_000369|0001_Angry|ZH|不就是你嘛，为什么要偷笑来。
+0001_000355|0001_Angry|ZH|我们乘船漂游了三峡，真是刺激。
+0001_000357|0001_Angry|ZH|我每个月打一次电话。
+0001_000356|0001_Angry|ZH|我喜欢“北京欢迎你”。
+0001_000367|0001_Angry|ZH|很快你上大学就用得到了。
+0001_000366|0001_Angry|ZH|他一定是一眼就被你迷住了。
+0001_000358|0001_Angry|ZH|沙尘暴好像给每个人都带来了麻烦！
+0001_000364|0001_Angry|ZH|谁你也不认识，我很乐意帮助你。
+0001_000370|0001_Angry|ZH|前几天我碰见了一件有趣的事儿。
+0001_000365|0001_Angry|ZH|我特别喜欢网球和登山。
+0001_000359|0001_Angry|ZH|就是这个意思，你又聪明又好看。
+0001_000361|0001_Angry|ZH|妇女节快乐。我永远爱你，妈妈。
+0001_000360|0001_Angry|ZH|个人收藏家！他们肯定有，
+0001_000362|0001_Angry|ZH|你每次谈恋爱都像现在这样。
+0001_000363|0001_Angry|ZH|周末的我，只忙着陪你。
+0001_000583|0001_Angry|ZH|拜托，别跟我提到笔记本电脑。
+0001_000597|0001_Angry|ZH|没有为什么就是要等我。
+0001_000568|0001_Angry|ZH|明天是星期天，我们去透透气吧。
+0001_000540|0001_Angry|ZH|上海现在是下午四点三十六分。
+0001_000554|0001_Angry|ZH|你昨天才买衣服，真是一购物狂。
+0001_000408|0001_Angry|ZH|让人看上去就感到宽广，气魄非凡。
+0001_000434|0001_Angry|ZH|他好像跟他的秘书有过一腿。
+0001_000420|0001_Angry|ZH|心动不如行动，我不太擅长卖萌。
+0001_000636|0001_Angry|ZH|门儿都没有，现在还不会。
+0001_000622|0001_Angry|ZH|沈阳明天有雷阵雨，多云转晴。
+0001_000623|0001_Angry|ZH|但梅花往往被很多人忽视。
+0001_000637|0001_Angry|ZH|我太喜欢听了，所以不断重复着听。
+0001_000421|0001_Angry|ZH|只要令人鼓舞的电影我都喜欢。
+0001_000435|0001_Angry|ZH|是啊，他的健康我总放心不下。
+0001_000409|0001_Angry|ZH|只要是我能玩好的，我都喜欢。
+0001_000555|0001_Angry|ZH|还要叫她起床，怎么会不早起。
+0001_000541|0001_Angry|ZH|别问了！说多了都是眼泪！
+0001_000569|0001_Angry|ZH|大部分都是用诱饵钓到的。
+0001_000596|0001_Angry|ZH|领带对男人来说真必不可少。
+0001_000582|0001_Angry|ZH|我最近正在努力练习棋艺。
+0001_000594|0001_Angry|ZH|我的直系亲属人数不多。
+0001_000580|0001_Angry|ZH|那不打扰你了，我不敢约出去。
+0001_000557|0001_Angry|ZH|我是萌萌哒，你是呆呆哒。
+0001_000543|0001_Angry|ZH|带女友出去好好吃上一顿。
+0001_000423|0001_Angry|ZH|她总是带香甜甜的微笑。
+0001_000437|0001_Angry|ZH|是没什么但是挺别扭的。
+0001_000609|0001_Angry|ZH|我们想等一个合适的时候。
+0001_000621|0001_Angry|ZH|很大，令人兴奋但是嘈杂。
+0001_000635|0001_Angry|ZH|自己保重，记得要常联系。
+0001_000634|0001_Angry|ZH|有充足的时间购物和观光。
+0001_000620|0001_Angry|ZH|没有找到你想删除的闹钟。
+0001_000608|0001_Angry|ZH|这句话的意义我不太明白。
+0001_000436|0001_Angry|ZH|真想不到，游泳竟有如此多的好处，
+0001_000422|0001_Angry|ZH|资料全都不见了。气死我了。
+0001_000542|0001_Angry|ZH|如果我滚远了就回不来了。
+0001_000556|0001_Angry|ZH|是的，我知道，患难见真情。
+0001_000581|0001_Angry|ZH|最近很冷，风又大。
+0001_000595|0001_Angry|ZH|贾尼斯突然兴奋地大叫起来。
+0001_000591|0001_Angry|ZH|请勿进入竹林。不让进。
+0001_000585|0001_Angry|ZH|我爱运动，但是对篮球玩得不多。
+0001_000552|0001_Angry|ZH|那就一会再说，我好害怕。
+0001_000546|0001_Angry|ZH|我饿啦，我想去吃点东西。
+0001_000426|0001_Angry|ZH|是的，你是个大块头，我是守门员。
+0001_000432|0001_Angry|ZH|听说你要去香港看你叔叔。
+0001_000624|0001_Angry|ZH|只要不违法，我还是想留下它。
+0001_000630|0001_Angry|ZH|绝对不可以走到湖的中央。
+0001_000618|0001_Angry|ZH|谁都有烦的时候。
+0001_000619|0001_Angry|ZH|是很棒的一款手机，性价比超级高。
+0001_000631|0001_Angry|ZH|晚安，么么哒，满天都是小星星。
+0001_000625|0001_Angry|ZH|然后再找一个音乐播放器，
+0001_000433|0001_Angry|ZH|你身上的每一点都吸引着我。
+0001_000427|0001_Angry|ZH|奏婚礼进行曲了，他们过来了。
+0001_000547|0001_Angry|ZH|为什么不，交朋友不分性别。
+0001_000553|0001_Angry|ZH|希望我有一天也可以去那里。
+0001_000584|0001_Angry|ZH|它是一个主要的空气污染物。
+0001_000590|0001_Angry|ZH|大约一个小时左右。
+0001_000586|0001_Angry|ZH|我不需要嗅觉，所以没有鼻子。
+0001_000592|0001_Angry|ZH|为了不让鱼吃掉诗人的身体。
+0001_000545|0001_Angry|ZH|我以为你们国家的人都是麻将高手。
+0001_000551|0001_Angry|ZH|也就是一大堆照片。
+0001_000579|0001_Angry|ZH|这个位置不错，下车。
+0001_000431|0001_Angry|ZH|这个镇上所有的人都喜欢扯闲话。
+0001_000425|0001_Angry|ZH|我也想去看可爱的熊猫。
+0001_000419|0001_Angry|ZH|以后不要喝那么多了，伤身体。
+0001_000633|0001_Angry|ZH|我想所有中国人都会打乒乓球。
+0001_000627|0001_Angry|ZH|是的，所以我永不喝它的。
+0001_000626|0001_Angry|ZH|我当然喜欢，我很注意颜面。
+0001_000632|0001_Angry|ZH|有些人划船，有的人在进行花草活动
+0001_000418|0001_Angry|ZH|我应该给女朋友买玫瑰花的。
+0001_000424|0001_Angry|ZH|今年应该是第二十七个教师节。
+0001_000430|0001_Angry|ZH|看呀，我们差不多就装饰好了。
+0001_000578|0001_Angry|ZH|好的。新的音乐厅有一场音乐会。
+0001_000550|0001_Angry|ZH|我已经习惯这种气候了。
+0001_000544|0001_Angry|ZH|炖肉一小时，剩余三十分钟十八秒。
+0001_000593|0001_Angry|ZH|那样的话你应该穿讲究一点。
+0001_000587|0001_Angry|ZH|雪下得真大，带着我去购物。
+0001_000523|0001_Angry|ZH|一套古瓷器。它真的很珍贵。
+0001_000537|0001_Angry|ZH|我们一起为他办个惊喜派对。
+0001_000494|0001_Angry|ZH|节食减肥很痛苦。
+0001_000480|0001_Angry|ZH|我的希望是工作到倒下的那一天。
+0001_000457|0001_Angry|ZH|我的表两点四十二。可是它有点快。
+0001_000443|0001_Angry|ZH|别不好意思。再多吃些鸡肉。
+0001_000696|0001_Angry|ZH|好好休息一下，这个小木棍叫梯。
+0001_000682|0001_Angry|ZH|我想要那种新款的美国兵款式。
+0001_000669|0001_Angry|ZH|家里有全自动洗衣机。
+0001_000655|0001_Angry|ZH|我支持你。它是需要重做。
+0001_000641|0001_Angry|ZH|在法国南部，气候常年舒适宜人。
+0001_000640|0001_Angry|ZH|还有聊天记录。
+0001_000654|0001_Angry|ZH|我刚从苏格兰回来。
+0001_000668|0001_Angry|ZH|下午好，蒂娜，我想我问错人了。
+0001_000683|0001_Angry|ZH|错过这村可就没这个店了。
+0001_000697|0001_Angry|ZH|这两块是唐朝不同时期铸造的。
+0001_000442|0001_Angry|ZH|不会这么凑巧吧！我也是十六。
+0001_000456|0001_Angry|ZH|做不了什么，通常我会保持沉默。
+0001_000481|0001_Angry|ZH|吝啬鬼！他每天还骑自行车上学！
+0001_000495|0001_Angry|ZH|听大自然的声响，就像听音乐一样！
+0001_000536|0001_Angry|ZH|软妹子生气会说：讨厌，不理你啦。
+0001_000522|0001_Angry|ZH|我是个学生，服务器响应超时。
+0001_000508|0001_Angry|ZH|我的性格就是冷静并且客观。
+0001_000534|0001_Angry|ZH|二零一六年十一月五号是星期六。
+0001_000520|0001_Angry|ZH|这样你就有时间挥拍打球了。
+0001_000483|0001_Angry|ZH|是不是依然觉得我很可爱。
+0001_000497|0001_Angry|ZH|不知道。或许一双新鞋。
+0001_000468|0001_Angry|ZH|你的同学把你给捉弄了吧。
+0001_000440|0001_Angry|ZH|还不太糟糕，但是得躺在床上。
+0001_000454|0001_Angry|ZH|没有找到蒸鱼的计时。
+0001_000681|0001_Angry|ZH|于是我就问她能不能连我的票买了。
+0001_000695|0001_Angry|ZH|别总是闲着，找点事情干。
+0001_000642|0001_Angry|ZH|哪天我也许得和他谈谈。
+0001_000656|0001_Angry|ZH|我在一个机械化农场做工程师。
+0001_000657|0001_Angry|ZH|有时内在美更加重要。
+0001_000643|0001_Angry|ZH|振作点儿，我看了屏幕显示！
+0001_000694|0001_Angry|ZH|很漂亮，不过人多拥挤。
+0001_000680|0001_Angry|ZH|我们相处得很好，仅此而已。
+0001_000455|0001_Angry|ZH|我们俩合不来，还经常吵架。
+0001_000441|0001_Angry|ZH|如果有我能帮忙的请告诉我。
+0001_000469|0001_Angry|ZH|不过我想星期五走，
+0001_000496|0001_Angry|ZH|现在，我仍然有点紧张。
+0001_000482|0001_Angry|ZH|是你最牵挂的那个女人。
+0001_000521|0001_Angry|ZH|你这个小气鬼，很不幸，非常少。
+0001_000535|0001_Angry|ZH|今天真凉快，我希望主队输掉。
+0001_000509|0001_Angry|ZH|我不会牺牲我的健康来换取金钱的。
+0001_000531|0001_Angry|ZH|时间对珍尼来说是没有用的。
+0001_000525|0001_Angry|ZH|她此刻失业了，你最好不要惹她。
+0001_000519|0001_Angry|ZH|让我再想想，真相已经上传了。
+0001_000486|0001_Angry|ZH|我喜欢几乎所有的运动，
+0001_000492|0001_Angry|ZH|是啊，我还是个乳臭未干的小记者。
+0001_000445|0001_Angry|ZH|我倒是有一个爱好――收藏古董。
+0001_000451|0001_Angry|ZH|很快，车就可自动开了。
+0001_000479|0001_Angry|ZH|对别人没有，而对我就有。
+0001_000684|0001_Angry|ZH|我曾经养过，我太高兴了。
+0001_000690|0001_Angry|ZH|每天晚上跟你互道晚安真幸福。
+0001_000647|0001_Angry|ZH|三个，两个儿子一个女儿。
+0001_000653|0001_Angry|ZH|我一直到清晨四点才到家，
+0001_000652|0001_Angry|ZH|我希望你能和我一起想派对点子。
+0001_000646|0001_Angry|ZH|赌博往往是个祸根，
+0001_000691|0001_Angry|ZH|我会在你的脸上画鬼脸。
+0001_000685|0001_Angry|ZH|他们将于今年夏天结婚。
+0001_000478|0001_Angry|ZH|这些颜色也不太适合你。
+0001_000450|0001_Angry|ZH|就经常去我们宿舍附近的酒吧。
+0001_000444|0001_Angry|ZH|很神奇的样子，我搞不懂为什么。
+0001_000493|0001_Angry|ZH|女孩的心思你别猜，但是我不用猜。
+0001_000487|0001_Angry|ZH|也许能帮助你把事情弄清楚。
+0001_000518|0001_Angry|ZH|你得先回答我，你最喜欢谁。
+0001_000524|0001_Angry|ZH|他在这次竞选活动中花了数百万，
+0001_000530|0001_Angry|ZH|冬天雨非常多。我不喜欢雨天。
+0001_000526|0001_Angry|ZH|带上你的家人，但是他有丑闻。
+0001_000532|0001_Angry|ZH|大概足够支持我生活三个月的。
+0001_000491|0001_Angry|ZH|你可以去问鹦鹉啊，鹦鹉会说话。
+0001_000485|0001_Angry|ZH|我要学习一下相关知识。
+0001_000452|0001_Angry|ZH|我昨天遇到马克，他看起来很忧郁。
+0001_000446|0001_Angry|ZH|别小看我这发型！我还蛮喜欢的。
+0001_000693|0001_Angry|ZH|许多白领都参加到这个游戏里面，
+0001_000687|0001_Angry|ZH|你看起来很高兴，眼睛闪闪发亮。
+0001_000650|0001_Angry|ZH|这个我相信，我是登山爱好者。
+0001_000644|0001_Angry|ZH|我是银灰色的，我都被你说饿了。
+0001_000678|0001_Angry|ZH|太棒了，我们下午可以在湖里划船。
+0001_000679|0001_Angry|ZH|但是有时夏天比其它季节更迷人。
+0001_000645|0001_Angry|ZH|文学和经济，我喜欢很多著作。
+0001_000651|0001_Angry|ZH|播放歌单收藏，脑筋可动得真快。
+0001_000686|0001_Angry|ZH|谢谢你的夸奖，鲍伯上年纪了。
+0001_000692|0001_Angry|ZH|我缺钱用，所以上星期把它当了。
+0001_000447|0001_Angry|ZH|你女儿和她妈妈长得很像。
+0001_000453|0001_Angry|ZH|感觉好温暖呀，好的，一会儿见。
+0001_000484|0001_Angry|ZH|多教我些东西我会更聪明。
+0001_000490|0001_Angry|ZH|我也知道自己是大嘴巴。
+0001_000533|0001_Angry|ZH|你绝对猜不到她准备要孩子了。
+0001_000527|0001_Angry|ZH|你在我的心里折腾好久了。
+0001_000700|0001_Angry|ZH|是很难听的脏话，主人可别学了。
+0001_000502|0001_Angry|ZH|小心脚下。人行道上有个坑。
+0001_000516|0001_Angry|ZH|我想那些应该是草莓的种子。
+0001_000489|0001_Angry|ZH|他们的配合值得我们学习
+0001_000476|0001_Angry|ZH|不是，他住在沃斯盾的老房子里。
+0001_000462|0001_Angry|ZH|那棒极了，其实心情挺不错。
+0001_000648|0001_Angry|ZH|祝你春节快乐，全家幸福安康。
+0001_000674|0001_Angry|ZH|我喜欢吃中餐。
+0001_000660|0001_Angry|ZH|这局我让你开，今天我不想错过。
+0001_000661|0001_Angry|ZH|我们休息一下喝杯咖啡。
+0001_000675|0001_Angry|ZH|尽管提意见，我会改正的。
+0001_000649|0001_Angry|ZH|是的，我刚撞到了桌子。
+0001_000463|0001_Angry|ZH|玛丽，你看来很喜欢挖苦我。
+0001_000477|0001_Angry|ZH|你说我们在芝加哥要待三天的。
+0001_000488|0001_Angry|ZH|我总是控制不了它。
+0001_000517|0001_Angry|ZH|太妙了，我想换一些日元。
+0001_000503|0001_Angry|ZH|我们还等什么
+0001_000529|0001_Angry|ZH|她和维克分手了，所以她申请转调。
+0001_000515|0001_Angry|ZH|其实我们前天已经分手了。
+0001_000501|0001_Angry|ZH|我也最喜欢你，不要开枪。我投降
+0001_000449|0001_Angry|ZH|得了吧，别这么胆小啦。
+0001_000461|0001_Angry|ZH|每天早晨都是我妈妈帮他系的。
+0001_000475|0001_Angry|ZH|如果你想要纹身，你去纹好了。
+0001_000688|0001_Angry|ZH|年轻人当然要承担责任，
+0001_000663|0001_Angry|ZH|旅行结束后我将休息一段时间。
+0001_000677|0001_Angry|ZH|我讨厌吃醋，偶是不懂，你懂。
+0001_000676|0001_Angry|ZH|这么多笑话，一天讲不完！
+0001_000662|0001_Angry|ZH|如果她拒绝我，我会死的。
+0001_000689|0001_Angry|ZH|不要乱问女孩子的年龄。
+0001_000474|0001_Angry|ZH|是的，请返还我的钱，谢谢。
+0001_000460|0001_Angry|ZH|自己的事情要自己做。
+0001_000448|0001_Angry|ZH|我只会斗斗地主什么的。
+0001_000500|0001_Angry|ZH|这样子比较有趣。
+0001_000514|0001_Angry|ZH|我曾在一家船运公司里面做过六年。
+0001_000528|0001_Angry|ZH|你转一个，我想学习下。
+0001_000510|0001_Angry|ZH|当然！他是我们大学的班长。
+0001_000504|0001_Angry|ZH|我还不会说其他外语，只会普通话。
+0001_000538|0001_Angry|ZH|以后我要经常来这儿爬山。
+0001_000464|0001_Angry|ZH|我也是，我还有点儿口渴。
+0001_000470|0001_Angry|ZH|你还真是考虑周到。
+0001_000458|0001_Angry|ZH|让我们看看哪一种球技比较好。
+0001_000699|0001_Angry|ZH|是的，真是名副其实。
+0001_000666|0001_Angry|ZH|每次你看到一些时尚衣物时，
+0001_000672|0001_Angry|ZH|等待你的指令，随时可为你效劳。
+0001_000673|0001_Angry|ZH|他对谁都那么友好。
+0001_000667|0001_Angry|ZH|你看上去比以前更漂亮了。
+0001_000698|0001_Angry|ZH|我喜欢你的黑衣服，你的尖牙真酷。
+0001_000459|0001_Angry|ZH|太棒了，我其实挺饿的。
+0001_000471|0001_Angry|ZH|那你就离市区很远了。
+0001_000465|0001_Angry|ZH|真是个好习惯，通常看书或消遣。
+0001_000539|0001_Angry|ZH|我是一名教师，你可是好眼光。
+0001_000505|0001_Angry|ZH|我只打算放松一下自己。
+0001_000511|0001_Angry|ZH|他讲的笑话让我笑个不停。
+0001_000507|0001_Angry|ZH|你知道，有时候病人会不讲理。
+0001_000513|0001_Angry|ZH|我还不知道你认识弗兰克。

filelists/esd.list.cleaned

@@ -0,0 +1,243 @@
+0001_000352|0001_Angry|ZH|英国的哲学家曾经说过'|_ y ing g uo d e zh e x ve j ia c eng j ing sh uo g uo ' _|0 1 1 2 2 5 5 2 2 2 2 1 1 2 2 1 1 1 1 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000353|0001_Angry|ZH|我老家在北京,哇塞!太精彩了.|_ w o l ao j ia z ai b ei j ing , w a s ai ! t ai j ing c ai l e . _|0 2 2 3 3 1 1 4 4 3 3 1 1 0 1 1 1 1 0 4 4 1 1 3 3 5 5 0 0|1 2 2 2 2 2 2 1 2 2 1 2 2 2 2 1 1
+0001_000351|0001_Angry|ZH|打远一看,它们的确很是美丽,|_ d a y van y i k an , t a m en d i q ve h en sh ir m ei l i , _|0 2 2 3 3 2 2 4 4 0 1 1 5 5 2 2 4 4 3 3 4 4 3 3 4 4 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 2 1 1
+0001_000354|0001_Angry|ZH|不管怎么说主队好象是志在夺魁.|_ b u g uan z en m e sh uo zh u d ui h ao x iang sh ir zh ir z ai d uo k ui . _|0 4 4 3 3 3 3 5 5 1 1 3 3 4 4 3 3 4 4 4 4 4 4 4 4 2 2 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000368|0001_Angry|ZH|我们意见不和,咱们去那儿玩吧.|_ w o m en y i j ian b u h e , z an m en q v n a EE er w an b a . _|0 3 3 5 5 4 4 4 4 4 4 2 2 0 2 2 5 5 4 4 4 4 2 2 2 2 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000369|0001_Angry|ZH|不就是你嘛,为什么要偷笑来.|_ b u j iu sh ir n i m a , w ei sh en m e y ao t ou x iao l ai . _|0 2 2 4 4 4 4 3 3 5 5 0 4 4 2 2 5 5 4 4 1 1 4 4 2 2 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000355|0001_Angry|ZH|我们乘船漂游了三峡,真是刺激.|_ w o m en ch eng ch uan p iao y ou l e s an x ia , zh en sh ir c i0 j i . _|0 3 3 5 5 2 2 2 2 1 1 2 2 5 5 1 1 2 2 0 1 1 4 4 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000357|0001_Angry|ZH|我每个月打一次电话.|_ w o m ei g e y ve d a y i c i0 d ian h ua . _|0 2 2 3 3 5 5 4 4 3 3 2 2 4 4 4 4 4 4 0 0|1 2 2 2 2 2 2 2 2 2 1 1
+0001_000356|0001_Angry|ZH|我喜欢'北京欢迎你'.|_ w o x i h uan ' b ei j ing h uan y ing n i ' . _|0 2 2 3 3 5 5 0 3 3 1 1 1 1 2 2 3 3 0 0 0|1 2 2 2 1 2 2 2 2 2 1 1 1
+0001_000367|0001_Angry|ZH|很快你上大学就用得到了.|_ h en k uai n i sh ang d a x ve j iu y ong d e d ao l e . _|0 3 3 4 4 3 3 4 4 4 4 2 2 4 4 4 4 2 2 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000366|0001_Angry|ZH|他一定是一眼就被你迷住了.|_ t a y i d ing sh ir y i y En j iu b ei n i m i zh u l e . _|0 1 1 2 2 4 4 4 4 4 4 3 3 4 4 4 4 3 3 2 2 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000358|0001_Angry|ZH|沙尘暴好像给每个人都带来了麻烦!|_ sh a ch en b ao h ao x iang g ei m ei g e r en d ou d ai l ai l e m a f an ! _|0 1 1 2 2 4 4 3 3 4 4 2 2 3 3 5 5 2 2 1 1 4 4 2 2 5 5 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000364|0001_Angry|ZH|谁你也不认识,我很乐意帮助你.|_ sh ui n i y E b u r en sh ir , w o h en l e y i b ang zh u n i . _|0 2 2 2 2 3 3 2 2 4 4 5 5 0 2 2 3 3 4 4 4 4 1 1 4 4 3 3 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000370|0001_Angry|ZH|前几天我碰见了一件有趣的事儿.|_ q ian j i t ian w o p eng j ian l e y i j ian y ou q v d e sh ir EE er . _|0 2 2 3 3 1 1 3 3 4 4 4 4 5 5 2 2 4 4 3 3 4 4 5 5 4 4 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000365|0001_Angry|ZH|我特别喜欢网球和登山.|_ w o t e b ie x i h uan w ang q iu h e d eng sh an . _|0 3 3 4 4 2 2 3 3 5 5 3 3 2 2 2 2 1 1 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000359|0001_Angry|ZH|就是这个意思,你又聪明又好看.|_ j iu sh ir zh e g e y i s i0 , n i y ou c ong m ing y ou h ao k an . _|0 4 4 4 4 4 4 5 5 4 4 5 5 0 3 3 4 4 1 1 5 5 4 4 3 3 4 4 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000361|0001_Angry|ZH|妇女节快乐.我永远爱你,妈妈.|_ f u n v j ie k uai l e . w o y ong y van AA ai n i , m a m a . _|0 4 4 3 3 2 2 4 4 4 4 0 3 3 2 2 3 3 4 4 3 3 0 1 1 5 5 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 2 2 1 1
+0001_000360|0001_Angry|ZH|个人收藏家!他们肯定有,|_ g e r en sh ou c ang j ia ! t a m en k en d ing y ou , _|0 4 4 2 2 1 1 2 2 1 1 0 1 1 5 5 3 3 4 4 3 3 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000362|0001_Angry|ZH|你每次谈恋爱都像现在这样.|_ n i m ei c i0 t an l ian AA ai d ou x iang x ian z ai zh e y ang . _|0 2 2 3 3 4 4 2 2 4 4 4 4 1 1 4 4 4 4 4 4 4 4 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000363|0001_Angry|ZH|周末的我,只忙着陪你.|_ zh ou m o d e w o , zh ir m ang zh e p ei n i . _|0 1 1 4 4 5 5 3 3 0 3 3 2 2 5 5 2 2 3 3 0 0|1 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000583|0001_Angry|ZH|拜托,别跟我提到笔记本电脑.|_ b ai t uo , b ie g en w o t i d ao b i j i b en d ian n ao . _|0 4 4 1 1 0 2 2 1 1 3 3 2 2 4 4 3 3 4 4 3 3 4 4 3 3 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000597|0001_Angry|ZH|没有为什么就是要等我.|_ m ei y ou w ei sh en m e j iu sh ir y ao d eng w o . _|0 2 2 3 3 4 4 2 2 5 5 4 4 4 4 4 4 2 2 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000568|0001_Angry|ZH|明天是星期天,我们去透透气吧.|_ m ing t ian sh ir x ing q i t ian , w o m en q v t ou t ou q i b a . _|0 2 2 1 1 4 4 1 1 1 1 1 1 0 3 3 5 5 4 4 4 4 5 5 4 4 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000540|0001_Angry|ZH|上海现在是下午四点三十六分.|_ sh ang h ai x ian z ai sh ir x ia w u s i0 d ian s an sh ir l iu f en . _|0 4 4 3 3 4 4 4 4 4 4 4 4 3 3 4 4 3 3 1 1 2 2 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000554|0001_Angry|ZH|你昨天才买衣服,真是一购物狂.|_ n i z uo t ian c ai m ai y i f u , zh en sh ir y i g ou w u k uang . _|0 3 3 2 2 1 1 2 2 3 3 1 1 5 5 0 1 1 4 4 2 2 4 4 4 4 2 2 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000408|0001_Angry|ZH|让人看上去就感到宽广,气魄非凡.|_ r ang r en k an sh ang q v j iu g an d ao k uan g uang , q i p o f ei f an . _|0 4 4 2 2 4 4 4 4 5 5 4 4 3 3 4 4 1 1 3 3 0 4 4 4 4 1 1 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000434|0001_Angry|ZH|他好像跟他的秘书有过一腿.|_ t a h ao x iang g en t a d e m i sh u y ou g uo y i t ui . _|0 1 1 3 3 4 4 1 1 1 1 5 5 4 4 1 1 3 3 5 5 4 4 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000420|0001_Angry|ZH|心动不如行动,我不太擅长卖萌.|_ x in d ong b u r u x ing d ong , w o b u t ai sh an ch ang m ai m eng . _|0 1 1 4 4 4 4 2 2 2 2 4 4 0 3 3 2 2 4 4 4 4 2 2 4 4 2 2 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000636|0001_Angry|ZH|门儿都没有,现在还不会.|_ m en EE er d ou m ei y ou , x ian z ai h ai b u h ui . _|0 2 2 2 2 1 1 2 2 3 3 0 4 4 4 4 2 2 2 2 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000622|0001_Angry|ZH|沈阳明天有雷阵雨,多云转晴.|_ sh en y ang m ing t ian y ou l ei zh en y v , d uo y vn zh uan q ing . _|0 3 3 2 2 2 2 1 1 3 3 2 2 4 4 3 3 0 1 1 2 2 3 3 2 2 0 0|1 2 2 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000623|0001_Angry|ZH|但梅花往往被很多人忽视.|_ d an m ei h ua w ang w ang b ei h en d uo r en h u sh ir . _|0 4 4 2 2 1 1 2 2 3 3 4 4 3 3 1 1 2 2 1 1 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000637|0001_Angry|ZH|我太喜欢听了,所以不断重复着听.|_ w o t ai x i h uan t ing l e , s uo y i b u d uan ch ong f u zh e t ing . _|0 3 3 4 4 3 3 5 5 1 1 5 5 0 2 2 3 3 2 2 4 4 2 2 4 4 5 5 1 1 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 2 1 1
+0001_000421|0001_Angry|ZH|只要令人鼓舞的电影我都喜欢.|_ zh ir y ao l ing r en g u w u d e d ian y ing w o d ou x i h uan . _|0 3 3 4 4 4 4 2 2 2 2 3 3 5 5 4 4 3 3 3 3 1 1 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000435|0001_Angry|ZH|是啊,他的健康我总放心不下.|_ sh ir AA a , t a d e j ian k ang w o z ong f ang x in b u x ia . _|0 4 4 5 5 0 1 1 5 5 4 4 1 1 2 2 3 3 4 4 1 1 2 2 5 5 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000409|0001_Angry|ZH|只要是我能玩好的,我都喜欢.|_ zh ir y ao sh ir w o n eng w an h ao d e , w o d ou x i h uan . _|0 3 3 4 4 4 4 3 3 2 2 2 2 3 3 5 5 0 3 3 1 1 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000555|0001_Angry|ZH|还要叫她起床,怎么会不早起.|_ h ai y ao j iao t a q i ch uang , z en m e h ui b u z ao q i . _|0 2 2 4 4 4 4 1 1 3 3 2 2 0 3 3 5 5 4 4 4 4 2 2 3 3 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000541|0001_Angry|ZH|别问了!说多了都是眼泪!|_ b ie w en l e ! sh uo d uo l e d ou sh ir y En l ei ! _|0 2 2 4 4 5 5 0 1 1 1 1 5 5 1 1 4 4 3 3 4 4 0 0|1 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000569|0001_Angry|ZH|大部分都是用诱饵钓到的.|_ d a b u f en d ou sh ir y ong y ou EE er d iao d ao d e . _|0 4 4 4 4 5 5 1 1 4 4 4 4 4 4 3 3 4 4 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000596|0001_Angry|ZH|领带对男人来说真必不可少.|_ l ing d ai d ui n an r en l ai sh uo zh en b i b u k e sh ao . _|0 3 3 4 4 4 4 2 2 2 2 2 2 1 1 1 1 4 4 4 4 2 2 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000582|0001_Angry|ZH|我最近正在努力练习棋艺.|_ w o z ui j in zh eng z ai n u l i l ian x i q i y i . _|0 3 3 4 4 4 4 4 4 4 4 3 3 4 4 4 4 2 2 2 2 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000594|0001_Angry|ZH|我的直系亲属人数不多.|_ w o d e zh ir x i q in sh u r en sh u b u d uo . _|0 3 3 5 5 2 2 4 4 1 1 3 3 2 2 4 4 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000580|0001_Angry|ZH|那不打扰你了,我不敢约出去.|_ n a b u d a r ao n i l e , w o b u g an y ve ch u q v . _|0 4 4 4 4 2 2 3 3 3 3 5 5 0 3 3 4 4 3 3 1 1 1 1 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000557|0001_Angry|ZH|我是萌萌哒,你是呆呆哒.|_ w o sh ir m eng m eng d a , n i sh ir d ai d ai d a . _|0 3 3 4 4 2 2 2 2 5 5 0 3 3 4 4 1 1 1 1 5 5 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000543|0001_Angry|ZH|带女友出去好好吃上一顿.|_ d ai n v y ou ch u q v h ao h ao ch ir sh ang y i d un . _|0 4 4 2 2 3 3 1 1 5 5 3 3 5 5 1 1 4 4 2 2 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000423|0001_Angry|ZH|她总是带香甜甜的微笑.|_ t a z ong sh ir d ai x iang t ian t ian d e w ei x iao . _|0 1 1 3 3 4 4 4 4 1 1 2 2 2 2 5 5 1 1 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000437|0001_Angry|ZH|是没什么但是挺别扭的.|_ sh ir m ei sh en m e d an sh ir t ing b ie n iu d e . _|0 4 4 2 2 2 2 5 5 4 4 4 4 3 3 4 4 5 5 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000609|0001_Angry|ZH|我们想等一个合适的时候.|_ w o m en x iang d eng y i g e h e sh ir d e sh ir h ou . _|0 3 3 5 5 2 2 3 3 2 2 5 5 2 2 4 4 5 5 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000621|0001_Angry|ZH|很大,令人兴奋但是嘈杂.|_ h en d a , l ing r en x ing f en d an sh ir c ao z a . _|0 3 3 4 4 0 4 4 2 2 1 1 4 4 4 4 4 4 2 2 2 2 0 0|1 2 2 1 2 2 2 2 2 2 2 2 1 1
+0001_000635|0001_Angry|ZH|自己保重,记得要常联系.|_ z i0 j i b ao zh ong , j i d e y ao ch ang l ian x i . _|0 4 4 3 3 3 3 4 4 0 4 4 5 5 4 4 2 2 2 2 4 4 0 0|1 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000634|0001_Angry|ZH|有充足的时间购物和观光.|_ y ou ch ong z u d e sh ir j ian g ou w u h e g uan g uang . _|0 3 3 1 1 2 2 5 5 2 2 1 1 4 4 4 4 2 2 1 1 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000620|0001_Angry|ZH|没有找到你想删除的闹钟.|_ m ei y ou zh ao d ao n i x iang sh an ch u d e n ao zh ong . _|0 2 2 3 3 3 3 4 4 2 2 3 3 1 1 2 2 5 5 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000608|0001_Angry|ZH|这句话的意义我不太明白.|_ zh e j v h ua d e y i y i w o b u t ai m ing b ai . _|0 4 4 4 4 4 4 5 5 4 4 4 4 3 3 2 2 4 4 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000436|0001_Angry|ZH|真想不到,游泳竟有如此多的好处,|_ zh en x iang b u d ao , y ou y ong j ing y ou r u c i0 d uo d e h ao ch u , _|0 1 1 3 3 5 5 4 4 0 2 2 3 3 4 4 3 3 2 2 3 3 1 1 5 5 3 3 4 4 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000422|0001_Angry|ZH|资料全都不见了.气死我了.|_ z i0 l iao q van d ou b u j ian l e . q i s i0 w o l e . _|0 1 1 4 4 2 2 1 1 2 2 4 4 5 5 0 4 4 3 3 3 3 5 5 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000542|0001_Angry|ZH|如果我滚远了就回不来了.|_ r u g uo w o g un y van l e j iu h ui b u l ai l e . _|0 2 2 3 3 3 3 2 2 3 3 5 5 4 4 2 2 5 5 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000556|0001_Angry|ZH|是的,我知道,患难见真情.|_ sh ir d e , w o zh ir d ao , h uan n an j ian zh en q ing . _|0 4 4 5 5 0 3 3 1 1 4 4 0 4 4 4 4 4 4 1 1 2 2 0 0|1 2 2 1 2 2 2 1 2 2 2 2 2 1 1
+0001_000581|0001_Angry|ZH|最近很冷,风又大.|_ z ui j in h en l eng , f eng y ou d a . _|0 4 4 4 4 2 2 3 3 0 1 1 4 4 4 4 0 0|1 2 2 2 2 1 2 2 2 1 1
+0001_000595|0001_Angry|ZH|贾尼斯突然兴奋地大叫起来.|_ j ia n i s i0 t u r an x ing f en d i d a j iao q i l ai . _|0 3 3 2 2 1 1 1 1 2 2 1 1 4 4 5 5 4 4 4 4 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000591|0001_Angry|ZH|请勿进入竹林.不让进.|_ q ing w u j in r u zh u l in . b u r ang j in . _|0 3 3 4 4 4 4 4 4 2 2 2 2 0 2 2 4 4 4 4 0 0|1 2 2 2 2 2 2 1 2 2 2 1 1
+0001_000585|0001_Angry|ZH|我爱运动,但是对篮球玩得不多.|_ w o AA ai y vn d ong , d an sh ir d ui l an q iu w an d e b u d uo . _|0 3 3 4 4 4 4 4 4 0 4 4 4 4 4 4 2 2 2 2 2 2 5 5 4 4 1 1 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 2 2 1 1
+0001_000552|0001_Angry|ZH|那就一会再说,我好害怕.|_ n a j iu y i h ui z ai sh uo , w o h ao h ai p a . _|0 4 4 4 4 2 2 4 4 4 4 1 1 0 2 2 3 3 4 4 4 4 0 0|1 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000546|0001_Angry|ZH|我饿啦,我想去吃点东西.|_ w o EE e l a , w o x iang q v ch ir d ian d ong x i . _|0 3 3 4 4 5 5 0 2 2 3 3 4 4 1 1 3 3 1 1 5 5 0 0|1 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000426|0001_Angry|ZH|是的,你是个大块头,我是守门员.|_ sh ir d e , n i sh ir g e d a k uai t ou , w o sh ir sh ou m en y van . _|0 4 4 5 5 0 3 3 4 4 5 5 4 4 4 4 2 2 0 3 3 4 4 3 3 2 2 2 2 0 0|1 2 2 1 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000432|0001_Angry|ZH|听说你要去香港看你叔叔.|_ t ing sh uo n i y ao q v x iang g ang k an n i sh u sh u . _|0 1 1 1 1 3 3 4 4 4 4 1 1 3 3 4 4 3 3 1 1 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000624|0001_Angry|ZH|只要不违法,我还是想留下它.|_ zh ir y ao b u w ei f a , w o h ai sh ir x iang l iu x ia t a . _|0 3 3 4 4 4 4 2 2 3 3 0 3 3 2 2 4 4 3 3 2 2 4 4 1 1 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000630|0001_Angry|ZH|绝对不可以走到湖的中央.|_ j ve d ui b u k e y i z ou d ao h u d e zh ong y ang . _|0 2 2 4 4 4 4 2 2 3 3 3 3 4 4 2 2 5 5 1 1 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000618|0001_Angry|ZH|谁都有烦的时候.|_ sh ui d ou y ou f an d e sh ir h ou . _|0 2 2 1 1 3 3 2 2 5 5 2 2 5 5 0 0|1 2 2 2 2 2 2 2 1 1
+0001_000619|0001_Angry|ZH|是很棒的一款手机,性价比超级高.|_ sh ir h en b ang d e y i k uan sh ou j i , x ing j ia b i ch ao j i g ao . _|0 4 4 3 3 4 4 5 5 4 4 3 3 3 3 1 1 0 4 4 4 4 3 3 1 1 2 2 1 1 0 0|1 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000631|0001_Angry|ZH|晚安,么么哒,满天都是小星星.|_ w an AA an , m e m e d a , m an t ian d ou sh ir x iao x ing x ing . _|0 3 3 1 1 0 5 5 5 5 5 5 0 3 3 1 1 1 1 4 4 3 3 1 1 5 5 0 0|1 2 2 1 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000625|0001_Angry|ZH|然后再找一个音乐播放器,|_ r an h ou z ai zh ao y i g e y in y ve b o f ang q i , _|0 2 2 4 4 4 4 3 3 2 2 5 5 1 1 4 4 1 1 4 4 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000433|0001_Angry|ZH|你身上的每一点都吸引着我.|_ n i sh en sh ang d e m ei y i d ian d ou x i y in zh e w o . _|0 3 3 1 1 5 5 5 5 3 3 4 4 3 3 1 1 1 1 3 3 5 5 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000427|0001_Angry|ZH|奏婚礼进行曲了,他们过来了.|_ z ou h un l i j in x ing q v l e , t a m en g uo l ai l e . _|0 4 4 1 1 3 3 4 4 2 2 1 1 5 5 0 1 1 5 5 4 4 5 5 5 5 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000547|0001_Angry|ZH|为什么不,交朋友不分性别.|_ w ei sh en m e b u , j iao p eng y ou b u f en x ing b ie . _|0 4 4 2 2 5 5 4 4 0 1 1 2 2 5 5 4 4 1 1 4 4 2 2 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000553|0001_Angry|ZH|希望我有一天也可以去那里.|_ x i w ang w o y ou y i t ian y E k e y i q v n a l i . _|0 1 1 4 4 2 2 3 3 4 4 1 1 3 3 2 2 3 3 4 4 4 4 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000584|0001_Angry|ZH|它是一个主要的空气污染物.|_ t a sh ir y i g e zh u y ao d e k ong q i w u r an w u . _|0 1 1 4 4 2 2 5 5 3 3 4 4 5 5 1 1 4 4 1 1 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000590|0001_Angry|ZH|大约一个小时左右.|_ d a y ve y i g e x iao sh ir z uo y ou . _|0 4 4 1 1 2 2 5 5 3 3 2 2 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000586|0001_Angry|ZH|我不需要嗅觉,所以没有鼻子.|_ w o b u x v y ao x iu j ve , s uo y i m ei y ou b i z i0 . _|0 3 3 4 4 1 1 4 4 4 4 2 2 0 2 2 3 3 2 2 3 3 2 2 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000592|0001_Angry|ZH|为了不让鱼吃掉诗人的身体.|_ w ei l e b u r ang y v ch ir d iao sh ir r en d e sh en t i . _|0 4 4 5 5 2 2 4 4 2 2 1 1 4 4 1 1 2 2 5 5 1 1 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000545|0001_Angry|ZH|我以为你们国家的人都是麻将高手.|_ w o y i w ei n i m en g uo j ia d e r en d ou sh ir m a j iang g ao sh ou . _|0 2 2 3 3 2 2 3 3 5 5 2 2 1 1 5 5 2 2 1 1 4 4 2 2 1 1 1 1 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000551|0001_Angry|ZH|也就是一大堆照片.|_ y E j iu sh ir y i d a d ui zh ao p ian . _|0 3 3 4 4 4 4 2 2 4 4 1 1 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000579|0001_Angry|ZH|这个位置不错,下车.|_ zh e g e w ei zh ir b u c uo , x ia ch e . _|0 4 4 5 5 4 4 5 5 2 2 4 4 0 4 4 1 1 0 0|1 2 2 2 2 2 2 1 2 2 1 1
+0001_000431|0001_Angry|ZH|这个镇上所有的人都喜欢扯闲话.|_ zh e g e zh en sh ang s uo y ou d e r en d ou x i h uan ch e x ian h ua . _|0 4 4 5 5 4 4 4 4 2 2 3 3 5 5 2 2 1 1 3 3 5 5 3 3 2 2 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000425|0001_Angry|ZH|我也想去看可爱的熊猫.|_ w o y E x iang q v k an k e AA ai d e x iong m ao . _|0 3 3 2 2 3 3 4 4 4 4 3 3 4 4 5 5 2 2 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000419|0001_Angry|ZH|以后不要喝那么多了,伤身体.|_ y i h ou b u y ao h e n a m e d uo l e , sh ang sh en t i . _|0 3 3 4 4 2 2 4 4 1 1 4 4 5 5 1 1 5 5 0 1 1 1 1 3 3 0 0|1 2 2 2 2 2 2 2 2 2 1 2 2 2 1 1
+0001_000633|0001_Angry|ZH|我想所有中国人都会打乒乓球.|_ w o x iang s uo y ou zh ong g uo r en d ou h ui d a p ing p ang q iu . _|0 2 2 3 3 2 2 3 3 1 1 2 2 2 2 1 1 4 4 3 3 1 1 1 1 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000627|0001_Angry|ZH|是的,所以我永不喝它的.|_ sh ir d e , s uo y i w o y ong b u h e t a d e . _|0 4 4 5 5 0 2 2 2 2 3 3 3 3 4 4 1 1 1 1 5 5 0 0|1 2 2 1 2 2 2 2 2 2 2 2 1 1
+0001_000626|0001_Angry|ZH|我当然喜欢,我很注意颜面.|_ w o d ang r an x i h uan , w o h en zh u y i y En m ian . _|0 3 3 1 1 2 2 3 3 5 5 0 2 2 3 3 4 4 4 4 2 2 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000632|0001_Angry|ZH|有些人划船,有的人在进行花草活动|_ y ou x ie r en h ua ch uan , y ou d e r en z ai j in x ing h ua c ao h uo d ong _|0 3 3 1 1 2 2 2 2 2 2 0 3 3 5 5 2 2 4 4 4 4 2 2 1 1 3 3 2 2 4 4 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 2 2 2 1
+0001_000418|0001_Angry|ZH|我应该给女朋友买玫瑰花的.|_ w o y ing g ai g ei n v p eng y ou m ai m ei g ui h ua d e . _|0 3 3 1 1 1 1 3 3 3 3 2 2 5 5 3 3 2 2 5 5 1 1 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000424|0001_Angry|ZH|今年应该是第二十七个教师节.|_ j in n ian y ing g ai sh ir d i EE er sh ir q i g e j iao sh ir j ie . _|0 1 1 2 2 1 1 1 1 4 4 4 4 4 4 2 2 1 1 5 5 4 4 1 1 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000430|0001_Angry|ZH|看呀,我们差不多就装饰好了.|_ k an y a , w o m en ch a b u d uo j iu zh uang sh ir h ao l e . _|0 4 4 5 5 0 3 3 5 5 4 4 5 5 1 1 4 4 1 1 4 4 3 3 5 5 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000578|0001_Angry|ZH|好的.新的音乐厅有一场音乐会.|_ h ao d e . x in d e y in y ve t ing y ou y i ch ang y in y ve h ui . _|0 3 3 5 5 0 1 1 5 5 1 1 4 4 1 1 3 3 4 4 3 3 1 1 4 4 4 4 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000550|0001_Angry|ZH|我已经习惯这种气候了.|_ w o y i j ing x i g uan zh e zh ong q i h ou l e . _|0 2 2 3 3 1 1 2 2 4 4 4 4 3 3 4 4 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000544|0001_Angry|ZH|炖肉一小时,剩余三十分钟十八秒.|_ d un r ou y i x iao sh ir , sh eng y v s an sh ir f en zh ong sh ir b a m iao . _|0 4 4 4 4 4 4 3 3 2 2 0 4 4 2 2 1 1 2 2 1 1 1 1 2 2 1 1 3 3 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 2 2 1 1
+0001_000593|0001_Angry|ZH|那样的话你应该穿讲究一点.|_ n a y ang d e h ua n i y ing g ai ch uan j iang j iu y i d ian . _|0 4 4 4 4 5 5 4 4 3 3 1 1 1 1 1 1 3 3 5 5 4 4 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000587|0001_Angry|ZH|雪下得真大,带着我去购物.|_ x ve x ia d e zh en d a , d ai zh e w o q v g ou w u . _|0 3 3 4 4 5 5 1 1 4 4 0 4 4 5 5 3 3 4 4 4 4 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000523|0001_Angry|ZH|一套古瓷器.它真的很珍贵.|_ y i t ao g u c i0 q i . t a zh en d e h en zh en g ui . _|0 2 2 4 4 3 3 2 2 4 4 0 1 1 1 1 5 5 3 3 1 1 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000537|0001_Angry|ZH|我们一起为他办个惊喜派对.|_ w o m en y i q i w ei t a b an g e j ing x i p ai d ui . _|0 3 3 5 5 4 4 3 3 4 4 1 1 4 4 5 5 1 1 3 3 4 4 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000494|0001_Angry|ZH|节食减肥很痛苦.|_ j ie sh ir j ian f ei h en t ong k u . _|0 2 2 2 2 3 3 2 2 3 3 4 4 3 3 0 0|1 2 2 2 2 2 2 2 1 1
+0001_000480|0001_Angry|ZH|我的希望是工作到倒下的那一天.|_ w o d e x i w ang sh ir g ong z uo d ao d ao x ia d e n a y i t ian . _|0 3 3 5 5 1 1 4 4 4 4 1 1 4 4 4 4 3 3 4 4 5 5 4 4 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000457|0001_Angry|ZH|我的表两点四十二.可是它有点快.|_ w o d e b iao l iang d ian s i0 sh ir EE er . k e sh ir t a y ou d ian k uai . _|0 3 3 5 5 3 3 2 2 3 3 4 4 2 2 4 4 0 3 3 4 4 1 1 2 2 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000443|0001_Angry|ZH|别不好意思.再多吃些鸡肉.|_ b ie b u h ao y i s i0 . z ai d uo ch ir x ie j i r ou . _|0 2 2 4 4 3 3 4 4 5 5 0 4 4 1 1 1 1 1 1 1 1 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000696|0001_Angry|ZH|好好休息一下,这个小木棍叫梯.|_ h ao h ao x iu x i y i x ia , zh e g e x iao m u g un j iao t i . _|0 2 2 3 3 1 1 5 5 2 2 4 4 0 4 4 5 5 3 3 4 4 4 4 4 4 1 1 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000682|0001_Angry|ZH|我想要那种新款的美国兵款式.|_ w o x iang y ao n a zh ong x in k uan d e m ei g uo b ing k uan sh ir . _|0 2 2 3 3 4 4 4 4 3 3 1 1 3 3 5 5 3 3 2 2 1 1 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000669|0001_Angry|ZH|家里有全自动洗衣机.|_ j ia l i y ou q van z i0 d ong x i y i j i . _|0 1 1 3 3 3 3 2 2 4 4 4 4 3 3 1 1 1 1 0 0|1 2 2 2 2 2 2 2 2 2 1 1
+0001_000655|0001_Angry|ZH|我支持你.它是需要重做.|_ w o zh ir ch ir n i . t a sh ir x v y ao zh ong z uo . _|0 3 3 1 1 2 2 3 3 0 1 1 4 4 1 1 4 4 4 4 4 4 0 0|1 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000641|0001_Angry|ZH|在法国南部,气候常年舒适宜人.|_ z ai f a g uo n an b u , q i h ou ch ang n ian sh u sh ir y i r en . _|0 4 4 3 3 2 2 2 2 4 4 0 4 4 4 4 2 2 2 2 1 1 4 4 2 2 2 2 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 2 1 1
+0001_000640|0001_Angry|ZH|还有聊天记录.|_ h ai y ou l iao t ian j i l u . _|0 2 2 3 3 2 2 1 1 4 4 4 4 0 0|1 2 2 2 2 2 2 1 1
+0001_000654|0001_Angry|ZH|我刚从苏格兰回来.|_ w o g ang c ong s u g e l an h ui l ai . _|0 3 3 1 1 2 2 1 1 2 2 2 2 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000668|0001_Angry|ZH|下午好,蒂娜,我想我问错人了.|_ x ia w u h ao , d i n a , w o x iang w o w en c uo r en l e . _|0 4 4 3 3 3 3 0 4 4 4 4 0 3 3 2 2 3 3 4 4 4 4 2 2 5 5 0 0|1 2 2 2 1 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000683|0001_Angry|ZH|错过这村可就没这个店了.|_ c uo g uo zh e c un k e j iu m ei zh e g e d ian l e . _|0 4 4 4 4 4 4 1 1 3 3 4 4 2 2 4 4 5 5 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000697|0001_Angry|ZH|这两块是唐朝不同时期铸造的.|_ zh e l iang k uai sh ir t ang ch ao b u t ong sh ir q i zh u z ao d e . _|0 4 4 3 3 4 4 4 4 2 2 2 2 4 4 2 2 2 2 1 1 4 4 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000442|0001_Angry|ZH|不会这么凑巧吧!我也是十六.|_ b u h ui zh e m e c ou q iao b a ! w o y E sh ir sh ir l iu . _|0 2 2 4 4 4 4 5 5 4 4 3 3 5 5 0 2 2 3 3 4 4 2 2 4 4 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000456|0001_Angry|ZH|做不了什么,通常我会保持沉默.|_ z uo b u l iao sh en m e , t ong ch ang w o h ui b ao ch ir ch en m o . _|0 4 4 5 5 3 3 2 2 5 5 0 1 1 2 2 3 3 4 4 3 3 2 2 2 2 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 2 1 1
+0001_000481|0001_Angry|ZH|吝啬鬼!他每天还骑自行车上学!|_ l in s e g ui ! t a m ei t ian h ai q i z i0 x ing ch e sh ang x ve ! _|0 4 4 4 4 3 3 0 1 1 3 3 1 1 2 2 2 2 4 4 2 2 1 1 4 4 2 2 0 0|1 2 2 2 1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000495|0001_Angry|ZH|听大自然的声响,就像听音乐一样!|_ t ing d a z i0 r an d e sh eng x iang , j iu x iang t ing y in y ve y i y ang ! _|0 1 1 4 4 4 4 2 2 5 5 1 1 3 3 0 4 4 4 4 1 1 1 1 4 4 2 2 4 4 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000536|0001_Angry|ZH|软妹子生气会说,讨厌,不理你啦.|_ r uan m ei z i0 sh eng q i h ui sh uo , t ao y En , b u l i n i l a . _|0 3 3 4 4 5 5 1 1 4 4 4 4 1 1 0 3 3 4 4 0 4 4 3 3 3 3 5 5 0 0|1 2 2 2 2 2 2 2 1 2 2 1 2 2 2 2 1 1
+0001_000522|0001_Angry|ZH|我是个学生,服务器响应超时.|_ w o sh ir g e x ve sh eng , f u w u q i x iang y ing ch ao sh ir . _|0 3 3 4 4 5 5 2 2 5 5 0 2 2 4 4 4 4 3 3 4 4 1 1 2 2 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000508|0001_Angry|ZH|我的性格就是冷静并且客观.|_ w o d e x ing g e j iu sh ir l eng j ing b ing q ie k e g uan . _|0 3 3 5 5 4 4 2 2 4 4 4 4 3 3 4 4 4 4 3 3 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000534|0001_Angry|ZH|二零一六年十一月五号是星期六.|_ EE er l ing y i l iu n ian sh ir y i y ve w u h ao sh ir x ing q i l iu . _|0 4 4 2 2 1 1 4 4 2 2 2 2 2 2 4 4 3 3 4 4 4 4 1 1 1 1 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000520|0001_Angry|ZH|这样你就有时间挥拍打球了.|_ zh e y ang n i j iu y ou sh ir j ian h ui p ai d a q iu l e . _|0 4 4 4 4 3 3 4 4 3 3 2 2 1 1 1 1 1 1 3 3 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000483|0001_Angry|ZH|是不是依然觉得我很可爱.|_ sh ir b u sh ir y i r an j ve d e w o h en k e AA ai . _|0 4 4 5 5 4 4 1 1 2 2 2 2 5 5 2 2 3 3 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000497|0001_Angry|ZH|不知道.或许一双新鞋.|_ b u zh ir d ao . h uo x v y i sh uang x in x ie . _|0 4 4 1 1 4 4 0 4 4 3 3 4 4 1 1 1 1 2 2 0 0|1 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000468|0001_Angry|ZH|你的同学把你给捉弄了吧.|_ n i d e t ong x ve b a n i g ei zh uo n ong l e b a . _|0 3 3 5 5 2 2 2 2 3 3 2 2 3 3 1 1 4 4 5 5 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000440|0001_Angry|ZH|还不太糟糕,但是得躺在床上.|_ h ai b u t ai z ao g ao , d an sh ir d e t ang z ai ch uang sh ang . _|0 2 2 2 2 4 4 1 1 1 1 0 4 4 4 4 5 5 3 3 4 4 2 2 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000454|0001_Angry|ZH|没有找到蒸鱼的计时.|_ m ei y ou zh ao d ao zh eng y v d e j i sh ir . _|0 2 2 3 3 3 3 4 4 1 1 2 2 5 5 4 4 2 2 0 0|1 2 2 2 2 2 2 2 2 2 1 1
+0001_000681|0001_Angry|ZH|于是我就问她能不能连我的票买了.|_ y v sh ir w o j iu w en t a n eng b u n eng l ian w o d e p iao m ai l e . _|0 2 2 4 4 3 3 4 4 4 4 1 1 2 2 4 4 2 2 2 2 3 3 5 5 4 4 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000695|0001_Angry|ZH|别总是闲着,找点事情干.|_ b ie z ong sh ir x ian zh e , zh ao d ian sh ir q ing g an . _|0 2 2 3 3 4 4 2 2 5 5 0 2 2 3 3 4 4 5 5 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000642|0001_Angry|ZH|哪天我也许得和他谈谈.|_ n a t ian w o y E x v d e h e t a t an t an . _|0 3 3 1 1 3 3 2 2 3 3 5 5 2 2 1 1 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000656|0001_Angry|ZH|我在一个机械化农场做工程师.|_ w o z ai y i g e j i x ie h ua n ong ch ang z uo g ong ch eng sh ir . _|0 3 3 4 4 2 2 5 5 1 1 4 4 4 4 2 2 3 3 4 4 1 1 2 2 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000657|0001_Angry|ZH|有时内在美更加重要.|_ y ou sh ir n ei z ai m ei g eng j ia zh ong y ao . _|0 3 3 2 2 4 4 4 4 3 3 4 4 1 1 4 4 4 4 0 0|1 2 2 2 2 2 2 2 2 2 1 1
+0001_000643|0001_Angry|ZH|振作点儿,我看了屏幕显示!|_ zh en z uo d ian EE er , w o k an l e p ing m u x ian sh ir ! _|0 4 4 4 4 3 3 2 2 0 3 3 4 4 5 5 2 2 4 4 3 3 4 4 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000694|0001_Angry|ZH|很漂亮,不过人多拥挤.|_ h en p iao l iang , b u g uo r en d uo y ong j i . _|0 3 3 4 4 5 5 0 2 2 4 4 2 2 1 1 1 1 3 3 0 0|1 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000680|0001_Angry|ZH|我们相处得很好,仅此而已.|_ w o m en x iang ch u d e h en h ao , j in c i0 EE er y i . _|0 3 3 5 5 1 1 3 3 5 5 2 2 3 3 0 2 2 3 3 2 2 3 3 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000455|0001_Angry|ZH|我们俩合不来,还经常吵架.|_ w o m en l ia h e b u l ai , h ai j ing ch ang ch ao j ia . _|0 3 3 5 5 3 3 2 2 5 5 2 2 0 2 2 1 1 2 2 3 3 4 4 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000441|0001_Angry|ZH|如果有我能帮忙的请告诉我.|_ r u g uo y ou w o n eng b ang m ang d e q ing g ao s u w o . _|0 2 2 3 3 2 2 3 3 2 2 1 1 2 2 5 5 3 3 4 4 5 5 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000469|0001_Angry|ZH|不过我想星期五走,|_ b u g uo w o x iang x ing q i w u z ou , _|0 2 2 4 4 2 2 3 3 1 1 1 1 3 3 3 3 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000496|0001_Angry|ZH|现在,我仍然有点紧张.|_ x ian z ai , w o r eng r an y ou d ian j in zh ang . _|0 4 4 4 4 0 3 3 2 2 2 2 2 2 3 3 3 3 1 1 0 0|1 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000482|0001_Angry|ZH|是你最牵挂的那个女人.|_ sh ir n i z ui q ian g ua d e n a g e n v r en . _|0 4 4 3 3 4 4 1 1 4 4 5 5 4 4 5 5 3 3 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000521|0001_Angry|ZH|你这个小气鬼,很不幸,非常少.|_ n i zh e g e x iao q i g ui , h en b u x ing , f ei ch ang sh ao . _|0 3 3 4 4 5 5 3 3 5 5 3 3 0 3 3 2 2 4 4 0 1 1 2 2 3 3 0 0|1 2 2 2 2 2 2 1 2 2 2 1 2 2 2 1 1
+0001_000535|0001_Angry|ZH|今天真凉快,我希望主队输掉.|_ j in t ian zh en l iang k uai , w o x i w ang zh u d ui sh u d iao . _|0 1 1 1 1 1 1 2 2 5 5 0 3 3 1 1 4 4 3 3 4 4 1 1 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000509|0001_Angry|ZH|我不会牺牲我的健康来换取金钱的.|_ w o b u h ui x i sh eng w o d e j ian k ang l ai h uan q v j in q ian d e . _|0 3 3 2 2 4 4 1 1 1 1 3 3 5 5 4 4 1 1 2 2 4 4 3 3 1 1 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000531|0001_Angry|ZH|时间对珍尼来说是没有用的.|_ sh ir j ian d ui zh en n i l ai sh uo sh ir m ei y ou y ong d e . _|0 2 2 1 1 4 4 1 1 2 2 2 2 1 1 4 4 2 2 3 3 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000525|0001_Angry|ZH|她此刻失业了,你最好不要惹她.|_ t a c i0 k e sh ir y E l e , n i z ui h ao b u y ao r e t a . _|0 1 1 3 3 4 4 1 1 4 4 5 5 0 3 3 4 4 3 3 2 2 4 4 3 3 1 1 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000519|0001_Angry|ZH|让我再想想,真相已经上传了.|_ r ang w o z ai x iang x iang , zh en x iang y i j ing sh ang ch uan l e . _|0 4 4 3 3 4 4 3 3 5 5 0 1 1 4 4 3 3 1 1 4 4 2 2 5 5 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000486|0001_Angry|ZH|我喜欢几乎所有的运动,|_ w o x i h uan j i h u s uo y ou d e y vn d ong , _|0 2 2 3 3 5 5 1 1 1 1 2 2 3 3 5 5 4 4 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000492|0001_Angry|ZH|是啊,我还是个乳臭未干的小记者.|_ sh ir AA a , w o h ai sh ir g e r u x iu w ei g an d e x iao j i zh e . _|0 4 4 5 5 0 3 3 2 2 4 4 5 5 3 3 4 4 4 4 1 1 5 5 3 3 4 4 3 3 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000445|0001_Angry|ZH|我倒是有一个爱好收藏古董.|_ w o d ao sh ir y ou y i g e AA ai h ao sh ou c ang g u d ong . _|0 3 3 4 4 4 4 3 3 2 2 5 5 4 4 4 4 1 1 2 2 2 2 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000451|0001_Angry|ZH|很快,车就可自动开了.|_ h en k uai , ch e j iu k e z i0 d ong k ai l e . _|0 3 3 4 4 0 1 1 4 4 3 3 4 4 4 4 1 1 5 5 0 0|1 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000479|0001_Angry|ZH|对别人没有,而对我就有.|_ d ui b ie r en m ei y ou , EE er d ui w o j iu y ou . _|0 4 4 2 2 2 2 2 2 3 3 0 2 2 4 4 3 3 4 4 3 3 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000684|0001_Angry|ZH|我曾经养过,我太高兴了.|_ w o c eng j ing y ang g uo , w o t ai g ao x ing l e . _|0 3 3 2 2 1 1 3 3 4 4 0 3 3 4 4 1 1 4 4 5 5 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000690|0001_Angry|ZH|每天晚上跟你互道晚安真幸福.|_ m ei t ian w an sh ang g en n i h u d ao w an AA an zh en x ing f u . _|0 3 3 1 1 3 3 4 4 1 1 3 3 4 4 4 4 3 3 1 1 1 1 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000647|0001_Angry|ZH|三个,两个儿子一个女儿.|_ s an g e , l iang g e EE er z i0 y i g e n v EE er . _|0 1 1 5 5 0 3 3 5 5 2 2 5 5 2 2 5 5 3 3 2 2 0 0|1 2 2 1 2 2 2 2 2 2 2 2 1 1
+0001_000653|0001_Angry|ZH|我一直到清晨四点才到家,|_ w o y i zh ir d ao q ing ch en s i0 d ian c ai d ao j ia , _|0 3 3 4 4 2 2 4 4 1 1 2 2 4 4 3 3 2 2 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000652|0001_Angry|ZH|我希望你能和我一起想派对点子.|_ w o x i w ang n i n eng h e w o y i q i x iang p ai d ui d ian z i0 . _|0 3 3 1 1 4 4 3 3 2 2 2 2 3 3 4 4 3 3 3 3 4 4 4 4 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000646|0001_Angry|ZH|赌博往往是个祸根,|_ d u b o w ang w ang sh ir g e h uo g en , _|0 3 3 2 2 2 2 3 3 4 4 5 5 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000691|0001_Angry|ZH|我会在你的脸上画鬼脸.|_ w o h ui z ai n i d e l ian sh ang h ua g ui l ian . _|0 3 3 4 4 4 4 3 3 5 5 3 3 5 5 4 4 2 2 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000685|0001_Angry|ZH|他们将于今年夏天结婚.|_ t a m en j iang y v j in n ian x ia t ian j ie h un . _|0 1 1 5 5 1 1 2 2 1 1 2 2 4 4 1 1 2 2 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000478|0001_Angry|ZH|这些颜色也不太适合你.|_ zh e x ie y En s e y E b u t ai sh ir h e n i . _|0 4 4 1 1 2 2 4 4 3 3 2 2 4 4 4 4 2 2 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000450|0001_Angry|ZH|就经常去我们宿舍附近的酒吧.|_ j iu j ing ch ang q v w o m en s u sh e f u j in d e j iu b a . _|0 4 4 1 1 2 2 4 4 3 3 5 5 4 4 4 4 4 4 4 4 5 5 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000444|0001_Angry|ZH|很神奇的样子,我搞不懂为什么.|_ h en sh en q i d e y ang z i0 , w o g ao b u d ong w ei sh en m e . _|0 3 3 2 2 2 2 5 5 4 4 5 5 0 3 3 3 3 5 5 3 3 4 4 2 2 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000493|0001_Angry|ZH|女孩的心思你别猜,但是我不用猜.|_ n v h ai d e x in s i0 n i b ie c ai , d an sh ir w o b u y ong c ai . _|0 3 3 2 2 5 5 1 1 5 5 3 3 2 2 1 1 0 4 4 4 4 3 3 2 2 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000487|0001_Angry|ZH|也许能帮助你把事情弄清楚.|_ y E x v n eng b ang zh u n i b a sh ir q ing n ong q ing ch u . _|0 2 2 3 3 2 2 1 1 4 4 2 2 3 3 4 4 5 5 4 4 1 1 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000518|0001_Angry|ZH|你得先回答我,你最喜欢谁.|_ n i d e x ian h ui d a w o , n i z ui x i h uan sh ui . _|0 3 3 5 5 1 1 2 2 2 2 3 3 0 3 3 4 4 3 3 5 5 2 2 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000524|0001_Angry|ZH|他在这次竞选活动中花了数百万,|_ t a z ai zh e c i0 j ing x van h uo d ong zh ong h ua l e sh u b ai w an , _|0 1 1 4 4 4 4 4 4 4 4 3 3 2 2 4 4 1 1 1 1 5 5 4 4 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000530|0001_Angry|ZH|冬天雨非常多.我不喜欢雨天.|_ d ong t ian y v f ei ch ang d uo . w o b u x i h uan y v t ian . _|0 1 1 1 1 3 3 1 1 2 2 1 1 0 3 3 4 4 3 3 5 5 3 3 1 1 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000526|0001_Angry|ZH|带上你的家人,但是他有丑闻.|_ d ai sh ang n i d e j ia r en , d an sh ir t a y ou ch ou w en . _|0 4 4 4 4 3 3 5 5 1 1 2 2 0 4 4 4 4 1 1 2 2 3 3 2 2 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000532|0001_Angry|ZH|大概足够支持我生活三个月的.|_ d a g ai z u g ou zh ir ch ir w o sh eng h uo s an g e y ve d e . _|0 4 4 4 4 2 2 4 4 1 1 2 2 3 3 1 1 2 2 1 1 5 5 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000491|0001_Angry|ZH|你可以去问鹦鹉啊,鹦鹉会说话.|_ n i k e y i q v w en y ing w u AA a , y ing w u h ui sh uo h ua . _|0 3 3 2 2 3 3 4 4 4 4 1 1 3 3 5 5 0 1 1 3 3 4 4 1 1 4 4 0 0|1 2 2 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000485|0001_Angry|ZH|我要学习一下相关知识.|_ w o y ao x ve x i y i x ia x iang g uan zh ir sh ir . _|0 3 3 4 4 2 2 2 2 2 2 4 4 1 1 1 1 1 1 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000452|0001_Angry|ZH|我昨天遇到马克,他看起来很忧郁.|_ w o z uo t ian y v d ao m a k e , t a k an q i l ai h en y ou y v . _|0 3 3 2 2 1 1 4 4 4 4 3 3 4 4 0 1 1 4 4 3 3 5 5 3 3 1 1 4 4 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000446|0001_Angry|ZH|别小看我这发型!我还蛮喜欢的.|_ b ie x iao k an w o zh e f a x ing ! w o h ai m an x i h uan d e . _|0 2 2 3 3 4 4 3 3 4 4 4 4 2 2 0 3 3 2 2 2 2 3 3 5 5 5 5 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000693|0001_Angry|ZH|许多白领都参加到这个游戏里面,|_ x v d uo b ai l ing d ou c an j ia d ao zh e g e y ou x i l i m ian , _|0 3 3 1 1 2 2 3 3 1 1 1 1 1 1 4 4 4 4 5 5 2 2 4 4 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000687|0001_Angry|ZH|你看起来很高兴,眼睛闪闪发亮.|_ n i k an q i l ai h en g ao x ing , y En j ing sh an sh an f a l iang . _|0 3 3 4 4 3 3 5 5 3 3 1 1 4 4 0 3 3 5 5 3 3 5 5 1 1 4 4 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000650|0001_Angry|ZH|这个我相信,我是登山爱好者.|_ zh e g e w o x iang x in , w o sh ir d eng sh an AA ai h ao zh e . _|0 4 4 5 5 3 3 1 1 4 4 0 3 3 4 4 1 1 1 1 4 4 4 4 3 3 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000644|0001_Angry|ZH|我是银灰色的,我都被你说饿了.|_ w o sh ir y in h ui s e d e , w o d ou b ei n i sh uo EE e l e . _|0 3 3 4 4 2 2 1 1 4 4 5 5 0 3 3 1 1 4 4 3 3 1 1 4 4 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000678|0001_Angry|ZH|太棒了,我们下午可以在湖里划船.|_ t ai b ang l e , w o m en x ia w u k e y i z ai h u l i h ua ch uan . _|0 4 4 4 4 5 5 0 3 3 5 5 4 4 3 3 2 2 3 3 4 4 2 2 5 5 2 2 2 2 0 0|1 2 2 2 1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000679|0001_Angry|ZH|但是有时夏天比其它季节更迷人.|_ d an sh ir y ou sh ir x ia t ian b i q i t a j i j ie g eng m i r en . _|0 4 4 4 4 3 3 2 2 4 4 1 1 3 3 2 2 1 1 4 4 2 2 4 4 2 2 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000645|0001_Angry|ZH|文学和经济,我喜欢很多著作.|_ w en x ve h e j ing j i , w o x i h uan h en d uo zh u z uo . _|0 2 2 2 2 2 2 1 1 4 4 0 2 2 3 3 5 5 3 3 1 1 4 4 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000651|0001_Angry|ZH|播放歌单收藏,脑筋可动得真快.|_ b o f ang g e d an sh ou c ang , n ao j in k e d ong d e zh en k uai . _|0 1 1 4 4 1 1 1 1 1 1 2 2 0 3 3 1 1 3 3 4 4 5 5 1 1 4 4 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000686|0001_Angry|ZH|谢谢你的夸奖,鲍伯上年纪了.|_ x ie x ie n i d e k ua j iang , b ao b o sh ang n ian j i l e . _|0 4 4 5 5 3 3 5 5 1 1 3 3 0 4 4 2 2 4 4 2 2 4 4 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000692|0001_Angry|ZH|我缺钱用,所以上星期把它当了.|_ w o q ve q ian y ong , s uo y i sh ang x ing q i b a t a d ang l e . _|0 3 3 1 1 2 2 4 4 0 2 2 3 3 4 4 1 1 1 1 3 3 1 1 1 1 5 5 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 2 2 1 1
+0001_000447|0001_Angry|ZH|你女儿和她妈妈长得很像.|_ n i n v EE er h e t a m a m a zh ang d e h en x iang . _|0 2 2 3 3 2 2 2 2 1 1 1 1 5 5 3 3 5 5 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000453|0001_Angry|ZH|感觉好温暖呀,好的,一会儿见.|_ g an j ve h ao w en n uan y a , h ao d e , y i h ui EE er j ian . _|0 3 3 2 2 3 3 1 1 3 3 5 5 0 3 3 5 5 0 2 2 4 4 5 5 4 4 0 0|1 2 2 2 2 2 2 1 2 2 1 2 2 2 2 1 1
+0001_000484|0001_Angry|ZH|多教我些东西我会更聪明.|_ d uo j iao w o x ie d ong x i w o h ui g eng c ong m ing . _|0 1 1 4 4 3 3 1 1 1 1 5 5 3 3 4 4 4 4 1 1 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000490|0001_Angry|ZH|我也知道自己是大嘴巴.|_ w o y E zh ir d ao z i0 j i sh ir d a z ui b a . _|0 2 2 3 3 1 1 4 4 4 4 3 3 4 4 4 4 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000533|0001_Angry|ZH|你绝对猜不到她准备要孩子了.|_ n i j ve d ui c ai b u d ao t a zh un b ei y ao h ai z i0 l e . _|0 3 3 2 2 4 4 1 1 2 2 4 4 1 1 3 3 4 4 4 4 2 2 5 5 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000527|0001_Angry|ZH|你在我的心里折腾好久了.|_ n i z ai w o d e x in l i zh e t eng h ao j iu l e . _|0 3 3 4 4 3 3 5 5 1 1 5 5 1 1 5 5 2 2 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000700|0001_Angry|ZH|是很难听的脏话,主人可别学了.|_ sh ir h en n an t ing d e z ang h ua , zh u r en k e b ie x ve l e . _|0 4 4 3 3 2 2 1 1 5 5 1 1 4 4 0 3 3 2 2 3 3 2 2 2 2 5 5 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000502|0001_Angry|ZH|小心脚下.人行道上有个坑.|_ x iao x in j iao x ia . r en x ing d ao sh ang y ou g e k eng . _|0 3 3 1 1 3 3 5 5 0 2 2 2 2 4 4 4 4 3 3 5 5 1 1 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000516|0001_Angry|ZH|我想那些应该是草莓的种子.|_ w o x iang n a x ie y ing g ai sh ir c ao m ei d e zh ong z i0 . _|0 2 2 3 3 4 4 1 1 1 1 1 1 4 4 3 3 2 2 5 5 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000489|0001_Angry|ZH|他们的配合值得我们学习|_ t a m en d e p ei h e zh ir d e w o m en x ve x i _|0 1 1 5 5 5 5 4 4 2 2 2 2 5 5 3 3 5 5 2 2 2 2 0|1 2 2 2 2 2 2 2 2 2 2 2 1
+0001_000476|0001_Angry|ZH|不是,他住在沃斯盾的老房子里.|_ b u sh ir , t a zh u z ai w o s i0 d un d e l ao f ang z i0 l i . _|0 2 2 4 4 0 1 1 4 4 4 4 4 4 1 1 4 4 5 5 3 3 2 2 5 5 3 3 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000462|0001_Angry|ZH|那棒极了,其实心情挺不错.|_ n a b ang j i l e , q i sh ir x in q ing t ing b u c uo . _|0 4 4 4 4 2 2 5 5 0 2 2 2 2 1 1 2 2 3 3 5 5 4 4 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000648|0001_Angry|ZH|祝你春节快乐,全家幸福安康.|_ zh u n i ch un j ie k uai l e , q van j ia x ing f u AA an k ang . _|0 4 4 3 3 1 1 2 2 4 4 4 4 0 2 2 1 1 4 4 5 5 1 1 1 1 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000674|0001_Angry|ZH|我喜欢吃中餐.|_ w o x i h uan ch ir zh ong c an . _|0 2 2 3 3 5 5 1 1 1 1 1 1 0 0|1 2 2 2 2 2 2 1 1
+0001_000660|0001_Angry|ZH|这局我让你开,今天我不想错过.|_ zh e j v w o r ang n i k ai , j in t ian w o b u x iang c uo g uo . _|0 4 4 2 2 3 3 4 4 3 3 1 1 0 1 1 1 1 3 3 4 4 3 3 4 4 4 4 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000661|0001_Angry|ZH|我们休息一下喝杯咖啡.|_ w o m en x iu x i y i x ia h e b ei k a f ei . _|0 3 3 5 5 1 1 5 5 2 2 4 4 1 1 1 1 1 1 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000675|0001_Angry|ZH|尽管提意见,我会改正的.|_ j in g uan t i y i j ian , w o h ui g ai zh eng d e . _|0 2 2 3 3 2 2 4 4 4 4 0 3 3 4 4 3 3 4 4 5 5 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000649|0001_Angry|ZH|是的,我刚撞到了桌子.|_ sh ir d e , w o g ang zh uang d ao l e zh uo z i0 . _|0 4 4 5 5 0 3 3 1 1 4 4 4 4 5 5 1 1 5 5 0 0|1 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000463|0001_Angry|ZH|玛丽,你看来很喜欢挖苦我.|_ m a l i , n i k an l ai h en x i h uan w a k u w o . _|0 3 3 4 4 0 3 3 4 4 2 2 2 2 3 3 5 5 1 1 5 5 3 3 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 1 1
+0001_000477|0001_Angry|ZH|你说我们在芝加哥要待三天的.|_ n i sh uo w o m en z ai zh ir j ia g e y ao d ai s an t ian d e . _|0 3 3 1 1 3 3 5 5 4 4 1 1 1 1 1 1 4 4 4 4 1 1 1 1 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000488|0001_Angry|ZH|我总是控制不了它.|_ w o z ong sh ir k ong zh ir b u l iao t a . _|0 2 2 3 3 4 4 4 4 4 4 4 4 3 3 1 1 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000517|0001_Angry|ZH|太妙了,我想换一些日元.|_ t ai m iao l e , w o x iang h uan y i x ie r ir y van . _|0 4 4 4 4 5 5 0 2 2 3 3 4 4 4 4 1 1 4 4 2 2 0 0|1 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000503|0001_Angry|ZH|我们还等什么|_ w o m en h ai d eng sh en m e _|0 3 3 5 5 2 2 3 3 2 2 5 5 0|1 2 2 2 2 2 2 1
+0001_000529|0001_Angry|ZH|她和维克分手了,所以她申请转调.|_ t a h e w ei k e f en sh ou l e , s uo y i t a sh en q ing zh uan d iao . _|0 1 1 2 2 2 2 4 4 1 1 3 3 5 5 0 2 2 3 3 1 1 1 1 3 3 3 3 4 4 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000515|0001_Angry|ZH|其实我们前天已经分手了.|_ q i sh ir w o m en q ian t ian y i j ing f en sh ou l e . _|0 2 2 2 2 3 3 5 5 2 2 1 1 3 3 1 1 1 1 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000501|0001_Angry|ZH|我也最喜欢你,不要开枪.我投降|_ w o y E z ui x i h uan n i , b u y ao k ai q iang . w o t ou x iang _|0 2 2 3 3 4 4 3 3 5 5 3 3 0 2 2 4 4 1 1 1 1 0 3 3 2 2 2 2 0|1 2 2 2 2 2 2 1 2 2 2 2 1 2 2 2 1
+0001_000449|0001_Angry|ZH|得了吧,别这么胆小啦.|_ d e l e b a , b ie zh e m e d an x iao l a . _|0 2 2 5 5 5 5 0 2 2 4 4 5 5 2 2 3 3 5 5 0 0|1 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000461|0001_Angry|ZH|每天早晨都是我妈妈帮他系的.|_ m ei t ian z ao ch en d ou sh ir w o m a m a b ang t a x i d e . _|0 3 3 1 1 3 3 2 2 1 1 4 4 3 3 1 1 5 5 1 1 1 1 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000475|0001_Angry|ZH|如果你想要纹身,你去纹好了.|_ r u g uo n i x iang y ao w en sh en , n i q v w en h ao l e . _|0 2 2 3 3 3 3 3 3 4 4 2 2 1 1 0 3 3 4 4 2 2 3 3 5 5 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000688|0001_Angry|ZH|年轻人当然要承担责任,|_ n ian q ing r en d ang r an y ao ch eng d an z e r en , _|0 2 2 1 1 2 2 1 1 2 2 4 4 2 2 1 1 2 2 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000663|0001_Angry|ZH|旅行结束后我将休息一段时间.|_ l v x ing j ie sh u h ou w o j iang x iu x i y i d uan sh ir j ian . _|0 3 3 2 2 2 2 4 4 4 4 3 3 1 1 1 1 5 5 2 2 4 4 2 2 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000677|0001_Angry|ZH|我讨厌吃醋,偶是不懂,你懂.|_ w o t ao y En ch ir c u , OO ou sh ir b u d ong , n i d ong . _|0 2 2 3 3 4 4 1 1 4 4 0 3 3 4 4 4 4 3 3 0 2 2 3 3 0 0|1 2 2 2 2 2 1 2 2 2 2 1 2 2 1 1
+0001_000676|0001_Angry|ZH|这么多笑话,一天讲不完!|_ zh e m e d uo x iao h ua , y i t ian j iang b u w an ! _|0 4 4 5 5 1 1 4 4 5 5 0 4 4 1 1 3 3 4 4 2 2 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000662|0001_Angry|ZH|如果她拒绝我,我会死的.|_ r u g uo t a j v j ve w o , w o h ui s i0 d e . _|0 2 2 3 3 1 1 4 4 2 2 3 3 0 3 3 4 4 3 3 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000689|0001_Angry|ZH|不要乱问女孩子的年龄.|_ b u y ao l uan w en n v h ai z i0 d e n ian l ing . _|0 2 2 4 4 4 4 4 4 3 3 2 2 5 5 5 5 2 2 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000474|0001_Angry|ZH|是的,请返还我的钱,谢谢.|_ sh ir d e , q ing f an h uan w o d e q ian , x ie x ie . _|0 4 4 5 5 0 2 2 3 3 2 2 3 3 5 5 2 2 0 4 4 5 5 0 0|1 2 2 1 2 2 2 2 2 2 1 2 2 1 1
+0001_000460|0001_Angry|ZH|自己的事情要自己做.|_ z i0 j i d e sh ir q ing y ao z i0 j i z uo . _|0 4 4 3 3 5 5 4 4 5 5 4 4 4 4 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 2 1 1
+0001_000448|0001_Angry|ZH|我只会斗斗地主什么的.|_ w o zh ir h ui d ou d ou d i zh u sh en m e d e . _|0 2 2 3 3 4 4 4 4 4 4 4 4 3 3 2 2 5 5 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000500|0001_Angry|ZH|这样子比较有趣.|_ zh e y ang z i0 b i j iao y ou q v . _|0 4 4 4 4 5 5 3 3 4 4 3 3 4 4 0 0|1 2 2 2 2 2 2 2 1 1
+0001_000514|0001_Angry|ZH|我曾在一家船运公司里面做过六年.|_ w o c eng z ai y i j ia ch uan y vn g ong s i0 l i m ian z uo g uo l iu n ian . _|0 3 3 2 2 4 4 4 4 1 1 2 2 4 4 1 1 1 1 3 3 4 4 4 4 5 5 4 4 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000528|0001_Angry|ZH|你转一个,我想学习下.|_ n i zh uan y i g e , w o x iang x ve x i x ia . _|0 2 2 3 3 2 2 5 5 0 2 2 3 3 2 2 2 2 4 4 0 0|1 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000510|0001_Angry|ZH|当然!他是我们大学的班长.|_ d ang r an ! t a sh ir w o m en d a x ve d e b an zh ang . _|0 1 1 2 2 0 1 1 4 4 3 3 5 5 4 4 2 2 5 5 1 1 3 3 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 1 1
+0001_000504|0001_Angry|ZH|我还不会说其他外语,只会普通话.|_ w o h ai b u h ui sh uo q i t a w ai y v , zh ir h ui p u t ong h ua . _|0 3 3 2 2 2 2 4 4 1 1 2 2 1 1 4 4 3 3 0 3 3 4 4 3 3 1 1 4 4 0 0|1 2 2 2 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000538|0001_Angry|ZH|以后我要经常来这儿爬山.|_ y i h ou w o y ao j ing ch ang l ai zh e EE er p a sh an . _|0 3 3 4 4 3 3 4 4 1 1 2 2 2 2 4 4 2 2 2 2 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000464|0001_Angry|ZH|我也是,我还有点儿口渴.|_ w o y E sh ir , w o h ai y ou d ian EE er k ou k e . _|0 2 2 3 3 4 4 0 3 3 2 2 2 2 3 3 2 2 2 2 3 3 0 0|1 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000470|0001_Angry|ZH|你还真是考虑周到.|_ n i h ai zh en sh ir k ao l v zh ou d ao . _|0 3 3 2 2 1 1 4 4 3 3 4 4 1 1 4 4 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000458|0001_Angry|ZH|让我们看看哪一种球技比较好.|_ r ang w o m en k an k an n a y i zh ong q iu j i b i j iao h ao . _|0 4 4 3 3 5 5 4 4 5 5 3 3 4 4 3 3 2 2 4 4 3 3 4 4 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000699|0001_Angry|ZH|是的,真是名副其实.|_ sh ir d e , zh en sh ir m ing f u q i sh ir . _|0 4 4 5 5 0 1 1 4 4 2 2 4 4 2 2 2 2 0 0|1 2 2 1 2 2 2 2 2 2 1 1
+0001_000666|0001_Angry|ZH|每次你看到一些时尚衣物时,|_ m ei c i0 n i k an d ao y i x ie sh ir sh ang y i w u sh ir , _|0 3 3 4 4 3 3 4 4 4 4 4 4 1 1 2 2 4 4 1 1 4 4 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000672|0001_Angry|ZH|等待你的指令,随时可为你效劳.|_ d eng d ai n i d e zh ir l ing , s ui sh ir k e w ei n i x iao l ao . _|0 3 3 4 4 3 3 5 5 3 3 4 4 0 2 2 2 2 3 3 4 4 3 3 4 4 2 2 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000673|0001_Angry|ZH|他对谁都那么友好.|_ t a d ui sh ui d ou n a m e y ou h ao . _|0 1 1 4 4 2 2 1 1 4 4 5 5 2 2 3 3 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000667|0001_Angry|ZH|你看上去比以前更漂亮了.|_ n i k an sh ang q v b i y i q ian g eng p iao l iang l e . _|0 3 3 4 4 4 4 5 5 2 2 3 3 2 2 4 4 4 4 5 5 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000698|0001_Angry|ZH|我喜欢你的黑衣服,你的尖牙真酷.|_ w o x i h uan n i d e h ei y i f u , n i d e j ian y a zh en k u . _|0 2 2 3 3 5 5 3 3 5 5 1 1 1 1 5 5 0 3 3 5 5 1 1 2 2 1 1 4 4 0 0|1 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000459|0001_Angry|ZH|太棒了,我其实挺饿的.|_ t ai b ang l e , w o q i sh ir t ing EE e d e . _|0 4 4 4 4 5 5 0 3 3 2 2 2 2 3 3 4 4 5 5 0 0|1 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000471|0001_Angry|ZH|那你就离市区很远了.|_ n a n i j iu l i sh ir q v h en y van l e . _|0 4 4 3 3 4 4 2 2 4 4 1 1 2 2 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 1 1
+0001_000465|0001_Angry|ZH|真是个好习惯,通常看书或消遣.|_ zh en sh ir g e h ao x i g uan , t ong ch ang k an sh u h uo x iao q ian . _|0 1 1 4 4 5 5 3 3 2 2 4 4 0 1 1 2 2 4 4 1 1 4 4 1 1 3 3 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000539|0001_Angry|ZH|我是一名教师,你可是好眼光.|_ w o sh ir y i m ing j iao sh ir , n i k e sh ir h ao y En g uang . _|0 3 3 4 4 4 4 2 2 4 4 1 1 0 2 2 3 3 4 4 2 2 3 3 1 1 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000505|0001_Angry|ZH|我只打算放松一下自己.|_ w o zh ir d a s uan f ang s ong y i x ia z i0 j i . _|0 2 2 3 3 3 3 5 5 4 4 1 1 2 2 4 4 4 4 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000511|0001_Angry|ZH|他讲的笑话让我笑个不停.|_ t a j iang d e x iao h ua r ang w o x iao g e b u t ing . _|0 1 1 3 3 5 5 4 4 5 5 4 4 3 3 4 4 5 5 4 4 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000507|0001_Angry|ZH|你知道,有时候病人会不讲理.|_ n i zh ir d ao , y ou sh ir h ou b ing r en h ui b u j iang l i . _|0 3 3 1 1 4 4 0 3 3 2 2 5 5 4 4 2 2 4 4 4 4 2 2 3 3 0 0|1 2 2 2 1 2 2 2 2 2 2 2 2 2 1 1
+0001_000513|0001_Angry|ZH|我还不知道你认识弗兰克.|_ w o h ai b u zh ir d ao n i r en sh ir f u l an k e . _|0 3 3 2 2 4 4 1 1 4 4 3 3 4 4 5 5 2 2 2 2 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1

filelists/train.list

@@ -0,0 +1,241 @@
+0001_000432|0001_Angry|ZH|听说你要去香港看你叔叔.|_ t ing sh uo n i y ao q v x iang g ang k an n i sh u sh u . _|0 1 1 1 1 3 3 4 4 4 4 1 1 3 3 4 4 3 3 1 1 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000557|0001_Angry|ZH|我是萌萌哒,你是呆呆哒.|_ w o sh ir m eng m eng d a , n i sh ir d ai d ai d a . _|0 3 3 4 4 2 2 2 2 5 5 0 3 3 4 4 1 1 1 1 5 5 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000621|0001_Angry|ZH|很大,令人兴奋但是嘈杂.|_ h en d a , l ing r en x ing f en d an sh ir c ao z a . _|0 3 3 4 4 0 4 4 2 2 1 1 4 4 4 4 4 4 2 2 2 2 0 0|1 2 2 1 2 2 2 2 2 2 2 2 1 1
+0001_000420|0001_Angry|ZH|心动不如行动,我不太擅长卖萌.|_ x in d ong b u r u x ing d ong , w o b u t ai sh an ch ang m ai m eng . _|0 1 1 4 4 4 4 2 2 2 2 4 4 0 3 3 2 2 4 4 4 4 2 2 4 4 2 2 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000569|0001_Angry|ZH|大部分都是用诱饵钓到的.|_ d a b u f en d ou sh ir y ong y ou EE er d iao d ao d e . _|0 4 4 4 4 5 5 1 1 4 4 4 4 4 4 3 3 4 4 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000556|0001_Angry|ZH|是的,我知道,患难见真情.|_ sh ir d e , w o zh ir d ao , h uan n an j ian zh en q ing . _|0 4 4 5 5 0 3 3 1 1 4 4 0 4 4 4 4 4 4 1 1 2 2 0 0|1 2 2 1 2 2 2 1 2 2 2 2 2 1 1
+0001_000486|0001_Angry|ZH|我喜欢几乎所有的运动,|_ w o x i h uan j i h u s uo y ou d e y vn d ong , _|0 2 2 3 3 5 5 1 1 1 1 2 2 3 3 5 5 4 4 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000656|0001_Angry|ZH|我在一个机械化农场做工程师.|_ w o z ai y i g e j i x ie h ua n ong ch ang z uo g ong ch eng sh ir . _|0 3 3 4 4 2 2 5 5 1 1 4 4 4 4 2 2 3 3 4 4 1 1 2 2 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000430|0001_Angry|ZH|看呀,我们差不多就装饰好了.|_ k an y a , w o m en ch a b u d uo j iu zh uang sh ir h ao l e . _|0 4 4 5 5 0 3 3 5 5 4 4 5 5 1 1 4 4 1 1 4 4 3 3 5 5 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000520|0001_Angry|ZH|这样你就有时间挥拍打球了.|_ zh e y ang n i j iu y ou sh ir j ian h ui p ai d a q iu l e . _|0 4 4 4 4 3 3 4 4 3 3 2 2 1 1 1 1 1 1 3 3 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000668|0001_Angry|ZH|下午好,蒂娜,我想我问错人了.|_ x ia w u h ao , d i n a , w o x iang w o w en c uo r en l e . _|0 4 4 3 3 3 3 0 4 4 4 4 0 3 3 2 2 3 3 4 4 4 4 2 2 5 5 0 0|1 2 2 2 1 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000624|0001_Angry|ZH|只要不违法,我还是想留下它.|_ zh ir y ao b u w ei f a , w o h ai sh ir x iang l iu x ia t a . _|0 3 3 4 4 4 4 2 2 3 3 0 3 3 2 2 4 4 3 3 2 2 4 4 1 1 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000642|0001_Angry|ZH|哪天我也许得和他谈谈.|_ n a t ian w o y E x v d e h e t a t an t an . _|0 3 3 1 1 3 3 2 2 3 3 5 5 2 2 1 1 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000360|0001_Angry|ZH|个人收藏家!他们肯定有,|_ g e r en sh ou c ang j ia ! t a m en k en d ing y ou , _|0 4 4 2 2 1 1 2 2 1 1 0 1 1 5 5 3 3 4 4 3 3 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000515|0001_Angry|ZH|其实我们前天已经分手了.|_ q i sh ir w o m en q ian t ian y i j ing f en sh ou l e . _|0 2 2 2 2 3 3 5 5 2 2 1 1 3 3 1 1 1 1 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000511|0001_Angry|ZH|他讲的笑话让我笑个不停.|_ t a j iang d e x iao h ua r ang w o x iao g e b u t ing . _|0 1 1 3 3 5 5 4 4 5 5 4 4 3 3 4 4 5 5 4 4 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000523|0001_Angry|ZH|一套古瓷器.它真的很珍贵.|_ y i t ao g u c i0 q i . t a zh en d e h en zh en g ui . _|0 2 2 4 4 3 3 2 2 4 4 0 1 1 1 1 5 5 3 3 1 1 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000633|0001_Angry|ZH|我想所有中国人都会打乒乓球.|_ w o x iang s uo y ou zh ong g uo r en d ou h ui d a p ing p ang q iu . _|0 2 2 3 3 2 2 3 3 1 1 2 2 2 2 1 1 4 4 3 3 1 1 1 1 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000666|0001_Angry|ZH|每次你看到一些时尚衣物时,|_ m ei c i0 n i k an d ao y i x ie sh ir sh ang y i w u sh ir , _|0 3 3 4 4 3 3 4 4 4 4 4 4 1 1 2 2 4 4 1 1 4 4 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000650|0001_Angry|ZH|这个我相信,我是登山爱好者.|_ zh e g e w o x iang x in , w o sh ir d eng sh an AA ai h ao zh e . _|0 4 4 5 5 3 3 1 1 4 4 0 3 3 4 4 1 1 1 1 4 4 4 4 3 3 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000580|0001_Angry|ZH|那不打扰你了,我不敢约出去.|_ n a b u d a r ao n i l e , w o b u g an y ve ch u q v . _|0 4 4 4 4 2 2 3 3 3 3 5 5 0 3 3 4 4 3 3 1 1 1 1 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000543|0001_Angry|ZH|带女友出去好好吃上一顿.|_ d ai n v y ou ch u q v h ao h ao ch ir sh ang y i d un . _|0 4 4 2 2 3 3 1 1 5 5 3 3 5 5 1 1 4 4 2 2 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000662|0001_Angry|ZH|如果她拒绝我,我会死的.|_ r u g uo t a j v j ve w o , w o h ui s i0 d e . _|0 2 2 3 3 1 1 4 4 2 2 3 3 0 3 3 4 4 3 3 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000536|0001_Angry|ZH|软妹子生气会说,讨厌,不理你啦.|_ r uan m ei z i0 sh eng q i h ui sh uo , t ao y En , b u l i n i l a . _|0 3 3 4 4 5 5 1 1 4 4 4 4 1 1 0 3 3 4 4 0 4 4 3 3 3 3 5 5 0 0|1 2 2 2 2 2 2 2 1 2 2 1 2 2 2 2 1 1
+0001_000451|0001_Angry|ZH|很快,车就可自动开了.|_ h en k uai , ch e j iu k e z i0 d ong k ai l e . _|0 3 3 4 4 0 1 1 4 4 3 3 4 4 4 4 1 1 5 5 0 0|1 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000476|0001_Angry|ZH|不是,他住在沃斯盾的老房子里.|_ b u sh ir , t a zh u z ai w o s i0 d un d e l ao f ang z i0 l i . _|0 2 2 4 4 0 1 1 4 4 4 4 4 4 1 1 4 4 5 5 3 3 2 2 5 5 3 3 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000544|0001_Angry|ZH|炖肉一小时,剩余三十分钟十八秒.|_ d un r ou y i x iao sh ir , sh eng y v s an sh ir f en zh ong sh ir b a m iao . _|0 4 4 4 4 4 4 3 3 2 2 0 4 4 2 2 1 1 2 2 1 1 1 1 2 2 1 1 3 3 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 2 2 1 1
+0001_000585|0001_Angry|ZH|我爱运动,但是对篮球玩得不多.|_ w o AA ai y vn d ong , d an sh ir d ui l an q iu w an d e b u d uo . _|0 3 3 4 4 4 4 4 4 0 4 4 4 4 4 4 2 2 2 2 2 2 5 5 4 4 1 1 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 2 2 1 1
+0001_000630|0001_Angry|ZH|绝对不可以走到湖的中央.|_ j ve d ui b u k e y i z ou d ao h u d e zh ong y ang . _|0 2 2 4 4 4 4 2 2 3 3 3 3 4 4 2 2 5 5 1 1 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000521|0001_Angry|ZH|你这个小气鬼,很不幸,非常少.|_ n i zh e g e x iao q i g ui , h en b u x ing , f ei ch ang sh ao . _|0 3 3 4 4 5 5 3 3 5 5 3 3 0 3 3 2 2 4 4 0 1 1 2 2 3 3 0 0|1 2 2 2 2 2 2 1 2 2 2 1 2 2 2 1 1
+0001_000427|0001_Angry|ZH|奏婚礼进行曲了,他们过来了.|_ z ou h un l i j in x ing q v l e , t a m en g uo l ai l e . _|0 4 4 1 1 3 3 4 4 2 2 1 1 5 5 0 1 1 5 5 4 4 5 5 5 5 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000522|0001_Angry|ZH|我是个学生,服务器响应超时.|_ w o sh ir g e x ve sh eng , f u w u q i x iang y ing ch ao sh ir . _|0 3 3 4 4 5 5 2 2 5 5 0 2 2 4 4 4 4 3 3 4 4 1 1 2 2 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000595|0001_Angry|ZH|贾尼斯突然兴奋地大叫起来.|_ j ia n i s i0 t u r an x ing f en d i d a j iao q i l ai . _|0 3 3 2 2 1 1 1 1 2 2 1 1 4 4 5 5 4 4 4 4 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000478|0001_Angry|ZH|这些颜色也不太适合你.|_ zh e x ie y En s e y E b u t ai sh ir h e n i . _|0 4 4 1 1 2 2 4 4 3 3 2 2 4 4 4 4 2 2 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000468|0001_Angry|ZH|你的同学把你给捉弄了吧.|_ n i d e t ong x ve b a n i g ei zh uo n ong l e b a . _|0 3 3 5 5 2 2 2 2 3 3 2 2 3 3 1 1 4 4 5 5 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000541|0001_Angry|ZH|别问了!说多了都是眼泪!|_ b ie w en l e ! sh uo d uo l e d ou sh ir y En l ei ! _|0 2 2 4 4 5 5 0 1 1 1 1 5 5 1 1 4 4 3 3 4 4 0 0|1 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000641|0001_Angry|ZH|在法国南部,气候常年舒适宜人.|_ z ai f a g uo n an b u , q i h ou ch ang n ian sh u sh ir y i r en . _|0 4 4 3 3 2 2 2 2 4 4 0 4 4 4 4 2 2 2 2 1 1 4 4 2 2 2 2 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 2 1 1
+0001_000465|0001_Angry|ZH|真是个好习惯,通常看书或消遣.|_ zh en sh ir g e h ao x i g uan , t ong ch ang k an sh u h uo x iao q ian . _|0 1 1 4 4 5 5 3 3 2 2 4 4 0 1 1 2 2 4 4 1 1 4 4 1 1 3 3 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000619|0001_Angry|ZH|是很棒的一款手机,性价比超级高.|_ sh ir h en b ang d e y i k uan sh ou j i , x ing j ia b i ch ao j i g ao . _|0 4 4 3 3 4 4 5 5 4 4 3 3 3 3 1 1 0 4 4 4 4 3 3 1 1 2 2 1 1 0 0|1 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000357|0001_Angry|ZH|我每个月打一次电话.|_ w o m ei g e y ve d a y i c i0 d ian h ua . _|0 2 2 3 3 5 5 4 4 3 3 2 2 4 4 4 4 4 4 0 0|1 2 2 2 2 2 2 2 2 2 1 1
+0001_000514|0001_Angry|ZH|我曾在一家船运公司里面做过六年.|_ w o c eng z ai y i j ia ch uan y vn g ong s i0 l i m ian z uo g uo l iu n ian . _|0 3 3 2 2 4 4 4 4 1 1 2 2 4 4 1 1 1 1 3 3 4 4 4 4 5 5 4 4 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000683|0001_Angry|ZH|错过这村可就没这个店了.|_ c uo g uo zh e c un k e j iu m ei zh e g e d ian l e . _|0 4 4 4 4 4 4 1 1 3 3 4 4 2 2 4 4 5 5 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000547|0001_Angry|ZH|为什么不,交朋友不分性别.|_ w ei sh en m e b u , j iao p eng y ou b u f en x ing b ie . _|0 4 4 2 2 5 5 4 4 0 1 1 2 2 5 5 4 4 1 1 4 4 2 2 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000673|0001_Angry|ZH|他对谁都那么友好.|_ t a d ui sh ui d ou n a m e y ou h ao . _|0 1 1 4 4 2 2 1 1 4 4 5 5 2 2 3 3 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000525|0001_Angry|ZH|她此刻失业了,你最好不要惹她.|_ t a c i0 k e sh ir y E l e , n i z ui h ao b u y ao r e t a . _|0 1 1 3 3 4 4 1 1 4 4 5 5 0 3 3 4 4 3 3 2 2 4 4 3 3 1 1 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000596|0001_Angry|ZH|领带对男人来说真必不可少.|_ l ing d ai d ui n an r en l ai sh uo zh en b i b u k e sh ao . _|0 3 3 4 4 4 4 2 2 2 2 2 2 1 1 1 1 4 4 4 4 2 2 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000409|0001_Angry|ZH|只要是我能玩好的,我都喜欢.|_ zh ir y ao sh ir w o n eng w an h ao d e , w o d ou x i h uan . _|0 3 3 4 4 4 4 3 3 2 2 2 2 3 3 5 5 0 3 3 1 1 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000590|0001_Angry|ZH|大约一个小时左右.|_ d a y ve y i g e x iao sh ir z uo y ou . _|0 4 4 1 1 2 2 5 5 3 3 2 2 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000686|0001_Angry|ZH|谢谢你的夸奖,鲍伯上年纪了.|_ x ie x ie n i d e k ua j iang , b ao b o sh ang n ian j i l e . _|0 4 4 5 5 3 3 5 5 1 1 3 3 0 4 4 2 2 4 4 2 2 4 4 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000369|0001_Angry|ZH|不就是你嘛,为什么要偷笑来.|_ b u j iu sh ir n i m a , w ei sh en m e y ao t ou x iao l ai . _|0 2 2 4 4 4 4 3 3 5 5 0 4 4 2 2 5 5 4 4 1 1 4 4 2 2 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000682|0001_Angry|ZH|我想要那种新款的美国兵款式.|_ w o x iang y ao n a zh ong x in k uan d e m ei g uo b ing k uan sh ir . _|0 2 2 3 3 4 4 4 4 3 3 1 1 3 3 5 5 3 3 2 2 1 1 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000516|0001_Angry|ZH|我想那些应该是草莓的种子.|_ w o x iang n a x ie y ing g ai sh ir c ao m ei d e zh ong z i0 . _|0 2 2 3 3 4 4 1 1 1 1 1 1 4 4 3 3 2 2 5 5 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000540|0001_Angry|ZH|上海现在是下午四点三十六分.|_ sh ang h ai x ian z ai sh ir x ia w u s i0 d ian s an sh ir l iu f en . _|0 4 4 3 3 4 4 4 4 4 4 4 4 3 3 4 4 3 3 1 1 2 2 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000661|0001_Angry|ZH|我们休息一下喝杯咖啡.|_ w o m en x iu x i y i x ia h e b ei k a f ei . _|0 3 3 5 5 1 1 5 5 2 2 4 4 1 1 1 1 1 1 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000578|0001_Angry|ZH|好的.新的音乐厅有一场音乐会.|_ h ao d e . x in d e y in y ve t ing y ou y i ch ang y in y ve h ui . _|0 3 3 5 5 0 1 1 5 5 1 1 4 4 1 1 3 3 4 4 3 3 1 1 4 4 4 4 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000418|0001_Angry|ZH|我应该给女朋友买玫瑰花的.|_ w o y ing g ai g ei n v p eng y ou m ai m ei g ui h ua d e . _|0 3 3 1 1 1 1 3 3 3 3 2 2 5 5 3 3 2 2 5 5 1 1 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000492|0001_Angry|ZH|是啊,我还是个乳臭未干的小记者.|_ sh ir AA a , w o h ai sh ir g e r u x iu w ei g an d e x iao j i zh e . _|0 4 4 5 5 0 3 3 2 2 4 4 5 5 3 3 4 4 4 4 1 1 5 5 3 3 4 4 3 3 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000433|0001_Angry|ZH|你身上的每一点都吸引着我.|_ n i sh en sh ang d e m ei y i d ian d ou x i y in zh e w o . _|0 3 3 1 1 5 5 5 5 3 3 4 4 3 3 1 1 1 1 3 3 5 5 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000471|0001_Angry|ZH|那你就离市区很远了.|_ n a n i j iu l i sh ir q v h en y van l e . _|0 4 4 3 3 4 4 2 2 4 4 1 1 2 2 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 1 1
+0001_000441|0001_Angry|ZH|如果有我能帮忙的请告诉我.|_ r u g uo y ou w o n eng b ang m ang d e q ing g ao s u w o . _|0 2 2 3 3 2 2 3 3 2 2 1 1 2 2 5 5 3 3 4 4 5 5 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000550|0001_Angry|ZH|我已经习惯这种气候了.|_ w o y i j ing x i g uan zh e zh ong q i h ou l e . _|0 2 2 3 3 1 1 2 2 4 4 4 4 3 3 4 4 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000496|0001_Angry|ZH|现在,我仍然有点紧张.|_ x ian z ai , w o r eng r an y ou d ian j in zh ang . _|0 4 4 4 4 0 3 3 2 2 2 2 2 2 3 3 3 3 1 1 0 0|1 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000462|0001_Angry|ZH|那棒极了,其实心情挺不错.|_ n a b ang j i l e , q i sh ir x in q ing t ing b u c uo . _|0 4 4 4 4 2 2 5 5 0 2 2 2 2 1 1 2 2 3 3 5 5 4 4 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000618|0001_Angry|ZH|谁都有烦的时候.|_ sh ui d ou y ou f an d e sh ir h ou . _|0 2 2 1 1 3 3 2 2 5 5 2 2 5 5 0 0|1 2 2 2 2 2 2 2 1 1
+0001_000623|0001_Angry|ZH|但梅花往往被很多人忽视.|_ d an m ei h ua w ang w ang b ei h en d uo r en h u sh ir . _|0 4 4 2 2 1 1 2 2 3 3 4 4 3 3 1 1 2 2 1 1 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000449|0001_Angry|ZH|得了吧,别这么胆小啦.|_ d e l e b a , b ie zh e m e d an x iao l a . _|0 2 2 5 5 5 5 0 2 2 4 4 5 5 2 2 3 3 5 5 0 0|1 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000423|0001_Angry|ZH|她总是带香甜甜的微笑.|_ t a z ong sh ir d ai x iang t ian t ian d e w ei x iao . _|0 1 1 3 3 4 4 4 4 1 1 2 2 2 2 5 5 1 1 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000367|0001_Angry|ZH|很快你上大学就用得到了.|_ h en k uai n i sh ang d a x ve j iu y ong d e d ao l e . _|0 3 3 4 4 3 3 4 4 4 4 2 2 4 4 4 4 2 2 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000652|0001_Angry|ZH|我希望你能和我一起想派对点子.|_ w o x i w ang n i n eng h e w o y i q i x iang p ai d ui d ian z i0 . _|0 3 3 1 1 4 4 3 3 2 2 2 2 3 3 4 4 3 3 3 3 4 4 4 4 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000655|0001_Angry|ZH|我支持你.它是需要重做.|_ w o zh ir ch ir n i . t a sh ir x v y ao zh ong z uo . _|0 3 3 1 1 2 2 3 3 0 1 1 4 4 1 1 4 4 4 4 4 4 0 0|1 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000517|0001_Angry|ZH|太妙了,我想换一些日元.|_ t ai m iao l e , w o x iang h uan y i x ie r ir y van . _|0 4 4 4 4 5 5 0 2 2 3 3 4 4 4 4 1 1 4 4 2 2 0 0|1 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000463|0001_Angry|ZH|玛丽,你看来很喜欢挖苦我.|_ m a l i , n i k an l ai h en x i h uan w a k u w o . _|0 3 3 4 4 0 3 3 4 4 2 2 2 2 3 3 5 5 1 1 5 5 3 3 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 1 1
+0001_000368|0001_Angry|ZH|我们意见不和,咱们去那儿玩吧.|_ w o m en y i j ian b u h e , z an m en q v n a EE er w an b a . _|0 3 3 5 5 4 4 4 4 4 4 2 2 0 2 2 5 5 4 4 4 4 2 2 2 2 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000657|0001_Angry|ZH|有时内在美更加重要.|_ y ou sh ir n ei z ai m ei g eng j ia zh ong y ao . _|0 3 3 2 2 4 4 4 4 3 3 4 4 1 1 4 4 4 4 0 0|1 2 2 2 2 2 2 2 2 2 1 1
+0001_000454|0001_Angry|ZH|没有找到蒸鱼的计时.|_ m ei y ou zh ao d ao zh eng y v d e j i sh ir . _|0 2 2 3 3 3 3 4 4 1 1 2 2 5 5 4 4 2 2 0 0|1 2 2 2 2 2 2 2 2 2 1 1
+0001_000679|0001_Angry|ZH|但是有时夏天比其它季节更迷人.|_ d an sh ir y ou sh ir x ia t ian b i q i t a j i j ie g eng m i r en . _|0 4 4 4 4 3 3 2 2 4 4 1 1 3 3 2 2 1 1 4 4 2 2 4 4 2 2 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000456|0001_Angry|ZH|做不了什么,通常我会保持沉默.|_ z uo b u l iao sh en m e , t ong ch ang w o h ui b ao ch ir ch en m o . _|0 4 4 5 5 3 3 2 2 5 5 0 1 1 2 2 3 3 4 4 3 3 2 2 2 2 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 2 1 1
+0001_000353|0001_Angry|ZH|我老家在北京,哇塞!太精彩了.|_ w o l ao j ia z ai b ei j ing , w a s ai ! t ai j ing c ai l e . _|0 2 2 3 3 1 1 4 4 3 3 1 1 0 1 1 1 1 0 4 4 1 1 3 3 5 5 0 0|1 2 2 2 2 2 2 1 2 2 1 2 2 2 2 1 1
+0001_000470|0001_Angry|ZH|你还真是考虑周到.|_ n i h ai zh en sh ir k ao l v zh ou d ao . _|0 3 3 2 2 1 1 4 4 3 3 4 4 1 1 4 4 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000542|0001_Angry|ZH|如果我滚远了就回不来了.|_ r u g uo w o g un y van l e j iu h ui b u l ai l e . _|0 2 2 3 3 3 3 2 2 3 3 5 5 4 4 2 2 5 5 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000592|0001_Angry|ZH|为了不让鱼吃掉诗人的身体.|_ w ei l e b u r ang y v ch ir d iao sh ir r en d e sh en t i . _|0 4 4 5 5 2 2 4 4 2 2 1 1 4 4 1 1 2 2 5 5 1 1 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000457|0001_Angry|ZH|我的表两点四十二.可是它有点快.|_ w o d e b iao l iang d ian s i0 sh ir EE er . k e sh ir t a y ou d ian k uai . _|0 3 3 5 5 3 3 2 2 3 3 4 4 2 2 4 4 0 3 3 4 4 1 1 2 2 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000464|0001_Angry|ZH|我也是,我还有点儿口渴.|_ w o y E sh ir , w o h ai y ou d ian EE er k ou k e . _|0 2 2 3 3 4 4 0 3 3 2 2 2 2 3 3 2 2 2 2 3 3 0 0|1 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000693|0001_Angry|ZH|许多白领都参加到这个游戏里面,|_ x v d uo b ai l ing d ou c an j ia d ao zh e g e y ou x i l i m ian , _|0 3 3 1 1 2 2 3 3 1 1 1 1 1 1 4 4 4 4 5 5 2 2 4 4 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000609|0001_Angry|ZH|我们想等一个合适的时候.|_ w o m en x iang d eng y i g e h e sh ir d e sh ir h ou . _|0 3 3 5 5 2 2 3 3 2 2 5 5 2 2 4 4 5 5 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000653|0001_Angry|ZH|我一直到清晨四点才到家,|_ w o y i zh ir d ao q ing ch en s i0 d ian c ai d ao j ia , _|0 3 3 4 4 2 2 4 4 1 1 2 2 4 4 3 3 2 2 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000685|0001_Angry|ZH|他们将于今年夏天结婚.|_ t a m en j iang y v j in n ian x ia t ian j ie h un . _|0 1 1 5 5 1 1 2 2 1 1 2 2 4 4 1 1 2 2 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000518|0001_Angry|ZH|你得先回答我,你最喜欢谁.|_ n i d e x ian h ui d a w o , n i z ui x i h uan sh ui . _|0 3 3 5 5 1 1 2 2 2 2 3 3 0 3 3 4 4 3 3 5 5 2 2 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000355|0001_Angry|ZH|我们乘船漂游了三峡,真是刺激.|_ w o m en ch eng ch uan p iao y ou l e s an x ia , zh en sh ir c i0 j i . _|0 3 3 5 5 2 2 2 2 1 1 2 2 5 5 1 1 2 2 0 1 1 4 4 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000584|0001_Angry|ZH|它是一个主要的空气污染物.|_ t a sh ir y i g e zh u y ao d e k ong q i w u r an w u . _|0 1 1 4 4 2 2 5 5 3 3 4 4 5 5 1 1 4 4 1 1 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000586|0001_Angry|ZH|我不需要嗅觉,所以没有鼻子.|_ w o b u x v y ao x iu j ve , s uo y i m ei y ou b i z i0 . _|0 3 3 4 4 1 1 4 4 4 4 2 2 0 2 2 3 3 2 2 3 3 2 2 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000488|0001_Angry|ZH|我总是控制不了它.|_ w o z ong sh ir k ong zh ir b u l iao t a . _|0 2 2 3 3 4 4 4 4 4 4 4 4 3 3 1 1 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000352|0001_Angry|ZH|英国的哲学家曾经说过'|_ y ing g uo d e zh e x ve j ia c eng j ing sh uo g uo ' _|0 1 1 2 2 5 5 2 2 2 2 1 1 2 2 1 1 1 1 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000660|0001_Angry|ZH|这局我让你开,今天我不想错过.|_ zh e j v w o r ang n i k ai , j in t ian w o b u x iang c uo g uo . _|0 4 4 2 2 3 3 4 4 3 3 1 1 0 1 1 1 1 3 3 4 4 3 3 4 4 4 4 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000408|0001_Angry|ZH|让人看上去就感到宽广,气魄非凡.|_ r ang r en k an sh ang q v j iu g an d ao k uan g uang , q i p o f ei f an . _|0 4 4 2 2 4 4 4 4 5 5 4 4 3 3 4 4 1 1 3 3 0 4 4 4 4 1 1 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000583|0001_Angry|ZH|拜托,别跟我提到笔记本电脑.|_ b ai t uo , b ie g en w o t i d ao b i j i b en d ian n ao . _|0 4 4 1 1 0 2 2 1 1 3 3 2 2 4 4 3 3 4 4 3 3 4 4 3 3 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000582|0001_Angry|ZH|我最近正在努力练习棋艺.|_ w o z ui j in zh eng z ai n u l i l ian x i q i y i . _|0 3 3 4 4 4 4 4 4 4 4 3 3 4 4 4 4 2 2 2 2 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000669|0001_Angry|ZH|家里有全自动洗衣机.|_ j ia l i y ou q van z i0 d ong x i y i j i . _|0 1 1 3 3 3 3 2 2 4 4 4 4 3 3 1 1 1 1 0 0|1 2 2 2 2 2 2 2 2 2 1 1
+0001_000359|0001_Angry|ZH|就是这个意思,你又聪明又好看.|_ j iu sh ir zh e g e y i s i0 , n i y ou c ong m ing y ou h ao k an . _|0 4 4 4 4 4 4 5 5 4 4 5 5 0 3 3 4 4 1 1 5 5 4 4 3 3 4 4 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000644|0001_Angry|ZH|我是银灰色的,我都被你说饿了.|_ w o sh ir y in h ui s e d e , w o d ou b ei n i sh uo EE e l e . _|0 3 3 4 4 2 2 1 1 4 4 5 5 0 3 3 1 1 4 4 3 3 1 1 4 4 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000469|0001_Angry|ZH|不过我想星期五走,|_ b u g uo w o x iang x ing q i w u z ou , _|0 2 2 4 4 2 2 3 3 1 1 1 1 3 3 3 3 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000358|0001_Angry|ZH|沙尘暴好像给每个人都带来了麻烦!|_ sh a ch en b ao h ao x iang g ei m ei g e r en d ou d ai l ai l e m a f an ! _|0 1 1 2 2 4 4 3 3 4 4 2 2 3 3 5 5 2 2 1 1 4 4 2 2 5 5 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000370|0001_Angry|ZH|前几天我碰见了一件有趣的事儿.|_ q ian j i t ian w o p eng j ian l e y i j ian y ou q v d e sh ir EE er . _|0 2 2 3 3 1 1 3 3 4 4 4 4 5 5 2 2 4 4 3 3 4 4 5 5 4 4 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000676|0001_Angry|ZH|这么多笑话,一天讲不完!|_ zh e m e d uo x iao h ua , y i t ian j iang b u w an ! _|0 4 4 5 5 1 1 4 4 5 5 0 4 4 1 1 3 3 4 4 2 2 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000597|0001_Angry|ZH|没有为什么就是要等我.|_ m ei y ou w ei sh en m e j iu sh ir y ao d eng w o . _|0 2 2 3 3 4 4 2 2 5 5 4 4 4 4 4 4 2 2 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000534|0001_Angry|ZH|二零一六年十一月五号是星期六.|_ EE er l ing y i l iu n ian sh ir y i y ve w u h ao sh ir x ing q i l iu . _|0 4 4 2 2 1 1 4 4 2 2 2 2 2 2 4 4 3 3 4 4 4 4 1 1 1 1 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000648|0001_Angry|ZH|祝你春节快乐,全家幸福安康.|_ zh u n i ch un j ie k uai l e , q van j ia x ing f u AA an k ang . _|0 4 4 3 3 1 1 2 2 4 4 4 4 0 2 2 1 1 4 4 5 5 1 1 1 1 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000537|0001_Angry|ZH|我们一起为他办个惊喜派对.|_ w o m en y i q i w ei t a b an g e j ing x i p ai d ui . _|0 3 3 5 5 4 4 3 3 4 4 1 1 4 4 5 5 1 1 3 3 4 4 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000510|0001_Angry|ZH|当然!他是我们大学的班长.|_ d ang r an ! t a sh ir w o m en d a x ve d e b an zh ang . _|0 1 1 2 2 0 1 1 4 4 3 3 5 5 4 4 2 2 5 5 1 1 3 3 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 1 1
+0001_000700|0001_Angry|ZH|是很难听的脏话,主人可别学了.|_ sh ir h en n an t ing d e z ang h ua , zh u r en k e b ie x ve l e . _|0 4 4 3 3 2 2 1 1 5 5 1 1 4 4 0 3 3 2 2 3 3 2 2 2 2 5 5 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000507|0001_Angry|ZH|你知道,有时候病人会不讲理.|_ n i zh ir d ao , y ou sh ir h ou b ing r en h ui b u j iang l i . _|0 3 3 1 1 4 4 0 3 3 2 2 5 5 4 4 2 2 4 4 4 4 2 2 3 3 0 0|1 2 2 2 1 2 2 2 2 2 2 2 2 2 1 1
+0001_000477|0001_Angry|ZH|你说我们在芝加哥要待三天的.|_ n i sh uo w o m en z ai zh ir j ia g e y ao d ai s an t ian d e . _|0 3 3 1 1 3 3 5 5 4 4 1 1 1 1 1 1 4 4 4 4 1 1 1 1 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000535|0001_Angry|ZH|今天真凉快,我希望主队输掉.|_ j in t ian zh en l iang k uai , w o x i w ang zh u d ui sh u d iao . _|0 1 1 1 1 1 1 2 2 5 5 0 3 3 1 1 4 4 3 3 4 4 1 1 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000440|0001_Angry|ZH|还不太糟糕,但是得躺在床上.|_ h ai b u t ai z ao g ao , d an sh ir d e t ang z ai ch uang sh ang . _|0 2 2 2 2 4 4 1 1 1 1 0 4 4 4 4 5 5 3 3 4 4 2 2 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000524|0001_Angry|ZH|他在这次竞选活动中花了数百万,|_ t a z ai zh e c i0 j ing x van h uo d ong zh ong h ua l e sh u b ai w an , _|0 1 1 4 4 4 4 4 4 4 4 3 3 2 2 4 4 1 1 1 1 5 5 4 4 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000627|0001_Angry|ZH|是的,所以我永不喝它的.|_ sh ir d e , s uo y i w o y ong b u h e t a d e . _|0 4 4 5 5 0 2 2 2 2 3 3 3 3 4 4 1 1 1 1 5 5 0 0|1 2 2 1 2 2 2 2 2 2 2 2 1 1
+0001_000503|0001_Angry|ZH|我们还等什么|_ w o m en h ai d eng sh en m e _|0 3 3 5 5 2 2 3 3 2 2 5 5 0|1 2 2 2 2 2 2 1
+0001_000581|0001_Angry|ZH|最近很冷,风又大.|_ z ui j in h en l eng , f eng y ou d a . _|0 4 4 4 4 2 2 3 3 0 1 1 4 4 4 4 0 0|1 2 2 2 2 1 2 2 2 1 1
+0001_000365|0001_Angry|ZH|我特别喜欢网球和登山.|_ w o t e b ie x i h uan w ang q iu h e d eng sh an . _|0 3 3 4 4 2 2 3 3 5 5 3 3 2 2 2 2 1 1 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000667|0001_Angry|ZH|你看上去比以前更漂亮了.|_ n i k an sh ang q v b i y i q ian g eng p iao l iang l e . _|0 3 3 4 4 4 4 5 5 2 2 3 3 2 2 4 4 4 4 5 5 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000421|0001_Angry|ZH|只要令人鼓舞的电影我都喜欢.|_ zh ir y ao l ing r en g u w u d e d ian y ing w o d ou x i h uan . _|0 3 3 4 4 4 4 2 2 2 2 3 3 5 5 4 4 3 3 3 3 1 1 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000424|0001_Angry|ZH|今年应该是第二十七个教师节.|_ j in n ian y ing g ai sh ir d i EE er sh ir q i g e j iao sh ir j ie . _|0 1 1 2 2 1 1 1 1 4 4 4 4 4 4 2 2 1 1 5 5 4 4 1 1 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000646|0001_Angry|ZH|赌博往往是个祸根,|_ d u b o w ang w ang sh ir g e h uo g en , _|0 3 3 2 2 2 2 3 3 4 4 5 5 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000485|0001_Angry|ZH|我要学习一下相关知识.|_ w o y ao x ve x i y i x ia x iang g uan zh ir sh ir . _|0 3 3 4 4 2 2 2 2 2 2 4 4 1 1 1 1 1 1 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000636|0001_Angry|ZH|门儿都没有,现在还不会.|_ m en EE er d ou m ei y ou , x ian z ai h ai b u h ui . _|0 2 2 2 2 1 1 2 2 3 3 0 4 4 4 4 2 2 2 2 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000663|0001_Angry|ZH|旅行结束后我将休息一段时间.|_ l v x ing j ie sh u h ou w o j iang x iu x i y i d uan sh ir j ian . _|0 3 3 2 2 2 2 4 4 4 4 3 3 1 1 1 1 5 5 2 2 4 4 2 2 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000351|0001_Angry|ZH|打远一看,它们的确很是美丽,|_ d a y van y i k an , t a m en d i q ve h en sh ir m ei l i , _|0 2 2 3 3 2 2 4 4 0 1 1 5 5 2 2 4 4 3 3 4 4 3 3 4 4 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 2 1 1
+0001_000680|0001_Angry|ZH|我们相处得很好,仅此而已.|_ w o m en x iang ch u d e h en h ao , j in c i0 EE er y i . _|0 3 3 5 5 1 1 3 3 5 5 2 2 3 3 0 2 2 3 3 2 2 3 3 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000422|0001_Angry|ZH|资料全都不见了.气死我了.|_ z i0 l iao q van d ou b u j ian l e . q i s i0 w o l e . _|0 1 1 4 4 2 2 1 1 2 2 4 4 5 5 0 4 4 3 3 3 3 5 5 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000675|0001_Angry|ZH|尽管提意见,我会改正的.|_ j in g uan t i y i j ian , w o h ui g ai zh eng d e . _|0 2 2 3 3 2 2 4 4 4 4 0 3 3 4 4 3 3 4 4 5 5 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000446|0001_Angry|ZH|别小看我这发型!我还蛮喜欢的.|_ b ie x iao k an w o zh e f a x ing ! w o h ai m an x i h uan d e . _|0 2 2 3 3 4 4 3 3 4 4 4 4 2 2 0 3 3 2 2 2 2 3 3 5 5 5 5 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000608|0001_Angry|ZH|这句话的意义我不太明白.|_ zh e j v h ua d e y i y i w o b u t ai m ing b ai . _|0 4 4 4 4 4 4 5 5 4 4 4 4 3 3 2 2 4 4 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000447|0001_Angry|ZH|你女儿和她妈妈长得很像.|_ n i n v EE er h e t a m a m a zh ang d e h en x iang . _|0 2 2 3 3 2 2 2 2 1 1 1 1 5 5 3 3 5 5 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000527|0001_Angry|ZH|你在我的心里折腾好久了.|_ n i z ai w o d e x in l i zh e t eng h ao j iu l e . _|0 3 3 4 4 3 3 5 5 1 1 5 5 1 1 5 5 2 2 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000504|0001_Angry|ZH|我还不会说其他外语,只会普通话.|_ w o h ai b u h ui sh uo q i t a w ai y v , zh ir h ui p u t ong h ua . _|0 3 3 2 2 2 2 4 4 1 1 2 2 1 1 4 4 3 3 0 3 3 4 4 3 3 1 1 4 4 0 0|1 2 2 2 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000692|0001_Angry|ZH|我缺钱用,所以上星期把它当了.|_ w o q ve q ian y ong , s uo y i sh ang x ing q i b a t a d ang l e . _|0 3 3 1 1 2 2 4 4 0 2 2 3 3 4 4 1 1 1 1 3 3 1 1 1 1 5 5 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 2 2 1 1
+0001_000554|0001_Angry|ZH|你昨天才买衣服,真是一购物狂.|_ n i z uo t ian c ai m ai y i f u , zh en sh ir y i g ou w u k uang . _|0 3 3 2 2 1 1 2 2 3 3 1 1 5 5 0 1 1 4 4 2 2 4 4 4 4 2 2 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000475|0001_Angry|ZH|如果你想要纹身,你去纹好了.|_ r u g uo n i x iang y ao w en sh en , n i q v w en h ao l e . _|0 2 2 3 3 3 3 3 3 4 4 2 2 1 1 0 3 3 4 4 2 2 3 3 5 5 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000493|0001_Angry|ZH|女孩的心思你别猜,但是我不用猜.|_ n v h ai d e x in s i0 n i b ie c ai , d an sh ir w o b u y ong c ai . _|0 3 3 2 2 5 5 1 1 5 5 3 3 2 2 1 1 0 4 4 4 4 3 3 2 2 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000691|0001_Angry|ZH|我会在你的脸上画鬼脸.|_ w o h ui z ai n i d e l ian sh ang h ua g ui l ian . _|0 3 3 4 4 4 4 3 3 5 5 3 3 5 5 4 4 2 2 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000688|0001_Angry|ZH|年轻人当然要承担责任,|_ n ian q ing r en d ang r an y ao ch eng d an z e r en , _|0 2 2 1 1 2 2 1 1 2 2 4 4 2 2 1 1 2 2 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000587|0001_Angry|ZH|雪下得真大,带着我去购物.|_ x ve x ia d e zh en d a , d ai zh e w o q v g ou w u . _|0 3 3 4 4 5 5 1 1 4 4 0 4 4 5 5 3 3 4 4 4 4 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000453|0001_Angry|ZH|感觉好温暖呀,好的,一会儿见.|_ g an j ve h ao w en n uan y a , h ao d e , y i h ui EE er j ian . _|0 3 3 2 2 3 3 1 1 3 3 5 5 0 3 3 5 5 0 2 2 4 4 5 5 4 4 0 0|1 2 2 2 2 2 2 1 2 2 1 2 2 2 2 1 1
+0001_000568|0001_Angry|ZH|明天是星期天,我们去透透气吧.|_ m ing t ian sh ir x ing q i t ian , w o m en q v t ou t ou q i b a . _|0 2 2 1 1 4 4 1 1 1 1 1 1 0 3 3 5 5 4 4 4 4 5 5 4 4 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000699|0001_Angry|ZH|是的,真是名副其实.|_ sh ir d e , zh en sh ir m ing f u q i sh ir . _|0 4 4 5 5 0 1 1 4 4 2 2 4 4 2 2 2 2 0 0|1 2 2 1 2 2 2 2 2 2 1 1
+0001_000530|0001_Angry|ZH|冬天雨非常多.我不喜欢雨天.|_ d ong t ian y v f ei ch ang d uo . w o b u x i h uan y v t ian . _|0 1 1 1 1 3 3 1 1 2 2 1 1 0 3 3 4 4 3 3 5 5 3 3 1 1 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000490|0001_Angry|ZH|我也知道自己是大嘴巴.|_ w o y E zh ir d ao z i0 j i sh ir d a z ui b a . _|0 2 2 3 3 1 1 4 4 4 4 3 3 4 4 4 4 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000363|0001_Angry|ZH|周末的我,只忙着陪你.|_ zh ou m o d e w o , zh ir m ang zh e p ei n i . _|0 1 1 4 4 5 5 3 3 0 3 3 2 2 5 5 2 2 3 3 0 0|1 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000484|0001_Angry|ZH|多教我些东西我会更聪明.|_ d uo j iao w o x ie d ong x i w o h ui g eng c ong m ing . _|0 1 1 4 4 3 3 1 1 1 1 5 5 3 3 4 4 4 4 1 1 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000674|0001_Angry|ZH|我喜欢吃中餐.|_ w o x i h uan ch ir zh ong c an . _|0 2 2 3 3 5 5 1 1 1 1 1 1 0 0|1 2 2 2 2 2 2 1 1
+0001_000645|0001_Angry|ZH|文学和经济,我喜欢很多著作.|_ w en x ve h e j ing j i , w o x i h uan h en d uo zh u z uo . _|0 2 2 2 2 2 2 1 1 4 4 0 2 2 3 3 5 5 3 3 1 1 4 4 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000526|0001_Angry|ZH|带上你的家人,但是他有丑闻.|_ d ai sh ang n i d e j ia r en , d an sh ir t a y ou ch ou w en . _|0 4 4 4 4 3 3 5 5 1 1 2 2 0 4 4 4 4 1 1 2 2 3 3 2 2 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000461|0001_Angry|ZH|每天早晨都是我妈妈帮他系的.|_ m ei t ian z ao ch en d ou sh ir w o m a m a b ang t a x i d e . _|0 3 3 1 1 3 3 2 2 1 1 4 4 3 3 1 1 5 5 1 1 1 1 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000634|0001_Angry|ZH|有充足的时间购物和观光.|_ y ou ch ong z u d e sh ir j ian g ou w u h e g uan g uang . _|0 3 3 1 1 2 2 5 5 2 2 1 1 4 4 4 4 2 2 1 1 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000508|0001_Angry|ZH|我的性格就是冷静并且客观.|_ w o d e x ing g e j iu sh ir l eng j ing b ing q ie k e g uan . _|0 3 3 5 5 4 4 2 2 4 4 4 4 3 3 4 4 4 4 3 3 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000555|0001_Angry|ZH|还要叫她起床,怎么会不早起.|_ h ai y ao j iao t a q i ch uang , z en m e h ui b u z ao q i . _|0 2 2 4 4 4 4 1 1 3 3 2 2 0 3 3 5 5 4 4 4 4 2 2 3 3 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000435|0001_Angry|ZH|是啊,他的健康我总放心不下.|_ sh ir AA a , t a d e j ian k ang w o z ong f ang x in b u x ia . _|0 4 4 5 5 0 1 1 5 5 4 4 1 1 2 2 3 3 4 4 1 1 2 2 5 5 0 0|1 2 2 1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000431|0001_Angry|ZH|这个镇上所有的人都喜欢扯闲话.|_ zh e g e zh en sh ang s uo y ou d e r en d ou x i h uan ch e x ian h ua . _|0 4 4 5 5 4 4 4 4 2 2 3 3 5 5 2 2 1 1 3 3 5 5 3 3 2 2 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000479|0001_Angry|ZH|对别人没有,而对我就有.|_ d ui b ie r en m ei y ou , EE er d ui w o j iu y ou . _|0 4 4 2 2 2 2 2 2 3 3 0 2 2 4 4 3 3 4 4 3 3 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000356|0001_Angry|ZH|我喜欢'北京欢迎你'.|_ w o x i h uan ' b ei j ing h uan y ing n i ' . _|0 2 2 3 3 5 5 0 3 3 1 1 1 1 2 2 3 3 0 0 0|1 2 2 2 1 2 2 2 2 2 1 1 1
+0001_000450|0001_Angry|ZH|就经常去我们宿舍附近的酒吧.|_ j iu j ing ch ang q v w o m en s u sh e f u j in d e j iu b a . _|0 4 4 1 1 2 2 4 4 3 3 5 5 4 4 4 4 4 4 4 4 5 5 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000551|0001_Angry|ZH|也就是一大堆照片.|_ y E j iu sh ir y i d a d ui zh ao p ian . _|0 3 3 4 4 4 4 2 2 4 4 1 1 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000495|0001_Angry|ZH|听大自然的声响,就像听音乐一样!|_ t ing d a z i0 r an d e sh eng x iang , j iu x iang t ing y in y ve y i y ang ! _|0 1 1 4 4 4 4 2 2 5 5 1 1 3 3 0 4 4 4 4 1 1 1 1 4 4 2 2 4 4 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000593|0001_Angry|ZH|那样的话你应该穿讲究一点.|_ n a y ang d e h ua n i y ing g ai ch uan j iang j iu y i d ian . _|0 4 4 4 4 5 5 4 4 3 3 1 1 1 1 1 1 3 3 5 5 4 4 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000538|0001_Angry|ZH|以后我要经常来这儿爬山.|_ y i h ou w o y ao j ing ch ang l ai zh e EE er p a sh an . _|0 3 3 4 4 3 3 4 4 1 1 2 2 2 2 4 4 2 2 2 2 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000681|0001_Angry|ZH|于是我就问她能不能连我的票买了.|_ y v sh ir w o j iu w en t a n eng b u n eng l ian w o d e p iao m ai l e . _|0 2 2 4 4 3 3 4 4 4 4 1 1 2 2 4 4 2 2 2 2 3 3 5 5 4 4 3 3 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000489|0001_Angry|ZH|他们的配合值得我们学习|_ t a m en d e p ei h e zh ir d e w o m en x ve x i _|0 1 1 5 5 5 5 4 4 2 2 2 2 5 5 3 3 5 5 2 2 2 2 0|1 2 2 2 2 2 2 2 2 2 2 2 1
+0001_000501|0001_Angry|ZH|我也最喜欢你,不要开枪.我投降|_ w o y E z ui x i h uan n i , b u y ao k ai q iang . w o t ou x iang _|0 2 2 3 3 4 4 3 3 5 5 3 3 0 2 2 4 4 1 1 1 1 0 3 3 2 2 2 2 0|1 2 2 2 2 2 2 1 2 2 2 2 1 2 2 2 1
+0001_000480|0001_Angry|ZH|我的希望是工作到倒下的那一天.|_ w o d e x i w ang sh ir g ong z uo d ao d ao x ia d e n a y i t ian . _|0 3 3 5 5 1 1 4 4 4 4 1 1 4 4 4 4 3 3 4 4 5 5 4 4 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000696|0001_Angry|ZH|好好休息一下,这个小木棍叫梯.|_ h ao h ao x iu x i y i x ia , zh e g e x iao m u g un j iao t i . _|0 2 2 3 3 1 1 5 5 2 2 4 4 0 4 4 5 5 3 3 4 4 4 4 4 4 1 1 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000509|0001_Angry|ZH|我不会牺牲我的健康来换取金钱的.|_ w o b u h ui x i sh eng w o d e j ian k ang l ai h uan q v j in q ian d e . _|0 3 3 2 2 4 4 1 1 1 1 3 3 5 5 4 4 1 1 2 2 4 4 3 3 1 1 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000448|0001_Angry|ZH|我只会斗斗地主什么的.|_ w o zh ir h ui d ou d ou d i zh u sh en m e d e . _|0 2 2 3 3 4 4 4 4 4 4 4 4 3 3 2 2 5 5 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000689|0001_Angry|ZH|不要乱问女孩子的年龄.|_ b u y ao l uan w en n v h ai z i0 d e n ian l ing . _|0 2 2 4 4 4 4 4 4 3 3 2 2 5 5 5 5 2 2 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000640|0001_Angry|ZH|还有聊天记录.|_ h ai y ou l iao t ian j i l u . _|0 2 2 3 3 2 2 1 1 4 4 4 4 0 0|1 2 2 2 2 2 2 1 1
+0001_000698|0001_Angry|ZH|我喜欢你的黑衣服,你的尖牙真酷.|_ w o x i h uan n i d e h ei y i f u , n i d e j ian y a zh en k u . _|0 2 2 3 3 5 5 3 3 5 5 1 1 1 1 5 5 0 3 3 5 5 1 1 2 2 1 1 4 4 0 0|1 2 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000632|0001_Angry|ZH|有些人划船,有的人在进行花草活动|_ y ou x ie r en h ua ch uan , y ou d e r en z ai j in x ing h ua c ao h uo d ong _|0 3 3 1 1 2 2 2 2 2 2 0 3 3 5 5 2 2 4 4 4 4 2 2 1 1 3 3 2 2 4 4 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 2 2 2 1
+0001_000513|0001_Angry|ZH|我还不知道你认识弗兰克.|_ w o h ai b u zh ir d ao n i r en sh ir f u l an k e . _|0 3 3 2 2 4 4 1 1 4 4 3 3 4 4 5 5 2 2 2 2 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000643|0001_Angry|ZH|振作点儿,我看了屏幕显示!|_ zh en z uo d ian EE er , w o k an l e p ing m u x ian sh ir ! _|0 4 4 4 4 3 3 2 2 0 3 3 4 4 5 5 2 2 4 4 3 3 4 4 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000528|0001_Angry|ZH|你转一个,我想学习下.|_ n i zh uan y i g e , w o x iang x ve x i x ia . _|0 2 2 3 3 2 2 5 5 0 2 2 3 3 2 2 2 2 4 4 0 0|1 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000626|0001_Angry|ZH|我当然喜欢,我很注意颜面.|_ w o d ang r an x i h uan , w o h en zh u y i y En m ian . _|0 3 3 1 1 2 2 3 3 5 5 0 2 2 3 3 4 4 4 4 2 2 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000491|0001_Angry|ZH|你可以去问鹦鹉啊,鹦鹉会说话.|_ n i k e y i q v w en y ing w u AA a , y ing w u h ui sh uo h ua . _|0 3 3 2 2 3 3 4 4 4 4 1 1 3 3 5 5 0 1 1 3 3 4 4 1 1 4 4 0 0|1 2 2 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000690|0001_Angry|ZH|每天晚上跟你互道晚安真幸福.|_ m ei t ian w an sh ang g en n i h u d ao w an AA an zh en x ing f u . _|0 3 3 1 1 3 3 4 4 1 1 3 3 4 4 4 4 3 3 1 1 1 1 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000482|0001_Angry|ZH|是你最牵挂的那个女人.|_ sh ir n i z ui q ian g ua d e n a g e n v r en . _|0 4 4 3 3 4 4 1 1 4 4 5 5 4 4 5 5 3 3 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000553|0001_Angry|ZH|希望我有一天也可以去那里.|_ x i w ang w o y ou y i t ian y E k e y i q v n a l i . _|0 1 1 4 4 2 2 3 3 4 4 1 1 3 3 2 2 3 3 4 4 4 4 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000533|0001_Angry|ZH|你绝对猜不到她准备要孩子了.|_ n i j ve d ui c ai b u d ao t a zh un b ei y ao h ai z i0 l e . _|0 3 3 2 2 4 4 1 1 2 2 4 4 1 1 3 3 4 4 4 4 2 2 5 5 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000579|0001_Angry|ZH|这个位置不错,下车.|_ zh e g e w ei zh ir b u c uo , x ia ch e . _|0 4 4 5 5 4 4 5 5 2 2 4 4 0 4 4 1 1 0 0|1 2 2 2 2 2 2 1 2 2 1 1
+0001_000500|0001_Angry|ZH|这样子比较有趣.|_ zh e y ang z i0 b i j iao y ou q v . _|0 4 4 4 4 5 5 3 3 4 4 3 3 4 4 0 0|1 2 2 2 2 2 2 2 1 1
+0001_000362|0001_Angry|ZH|你每次谈恋爱都像现在这样.|_ n i m ei c i0 t an l ian AA ai d ou x iang x ian z ai zh e y ang . _|0 2 2 3 3 4 4 2 2 4 4 4 4 1 1 4 4 4 4 4 4 4 4 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000505|0001_Angry|ZH|我只打算放松一下自己.|_ w o zh ir d a s uan f ang s ong y i x ia z i0 j i . _|0 2 2 3 3 3 3 5 5 4 4 1 1 2 2 4 4 4 4 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000434|0001_Angry|ZH|他好像跟他的秘书有过一腿.|_ t a h ao x iang g en t a d e m i sh u y ou g uo y i t ui . _|0 1 1 3 3 4 4 1 1 1 1 5 5 4 4 1 1 3 3 5 5 4 4 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000622|0001_Angry|ZH|沈阳明天有雷阵雨,多云转晴.|_ sh en y ang m ing t ian y ou l ei zh en y v , d uo y vn zh uan q ing . _|0 3 3 2 2 2 2 1 1 3 3 2 2 4 4 3 3 0 1 1 2 2 3 3 2 2 0 0|1 2 2 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000459|0001_Angry|ZH|太棒了,我其实挺饿的.|_ t ai b ang l e , w o q i sh ir t ing EE e d e . _|0 4 4 4 4 5 5 0 3 3 2 2 2 2 3 3 4 4 5 5 0 0|1 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000631|0001_Angry|ZH|晚安,么么哒,满天都是小星星.|_ w an AA an , m e m e d a , m an t ian d ou sh ir x iao x ing x ing . _|0 3 3 1 1 0 5 5 5 5 5 5 0 3 3 1 1 1 1 4 4 3 3 1 1 5 5 0 0|1 2 2 1 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000460|0001_Angry|ZH|自己的事情要自己做.|_ z i0 j i d e sh ir q ing y ao z i0 j i z uo . _|0 4 4 3 3 5 5 4 4 5 5 4 4 4 4 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 2 1 1
+0001_000354|0001_Angry|ZH|不管怎么说主队好象是志在夺魁.|_ b u g uan z en m e sh uo zh u d ui h ao x iang sh ir zh ir z ai d uo k ui . _|0 4 4 3 3 3 3 5 5 1 1 3 3 4 4 3 3 4 4 4 4 4 4 4 4 2 2 2 2 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000654|0001_Angry|ZH|我刚从苏格兰回来.|_ w o g ang c ong s u g e l an h ui l ai . _|0 3 3 1 1 2 2 1 1 2 2 2 2 2 2 5 5 0 0|1 2 2 2 2 2 2 2 2 1 1
+0001_000677|0001_Angry|ZH|我讨厌吃醋,偶是不懂,你懂.|_ w o t ao y En ch ir c u , OO ou sh ir b u d ong , n i d ong . _|0 2 2 3 3 4 4 1 1 4 4 0 3 3 4 4 4 4 3 3 0 2 2 3 3 0 0|1 2 2 2 2 2 1 2 2 2 2 1 2 2 1 1
+0001_000672|0001_Angry|ZH|等待你的指令,随时可为你效劳.|_ d eng d ai n i d e zh ir l ing , s ui sh ir k e w ei n i x iao l ao . _|0 3 3 4 4 3 3 5 5 3 3 4 4 0 2 2 2 2 3 3 4 4 3 3 4 4 2 2 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000436|0001_Angry|ZH|真想不到,游泳竟有如此多的好处,|_ zh en x iang b u d ao , y ou y ong j ing y ou r u c i0 d uo d e h ao ch u , _|0 1 1 3 3 5 5 4 4 0 2 2 3 3 4 4 3 3 2 2 3 3 1 1 5 5 3 3 4 4 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000635|0001_Angry|ZH|自己保重,记得要常联系.|_ z i0 j i b ao zh ong , j i d e y ao ch ang l ian x i . _|0 4 4 3 3 3 3 4 4 0 4 4 5 5 4 4 2 2 2 2 4 4 0 0|1 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000458|0001_Angry|ZH|让我们看看哪一种球技比较好.|_ r ang w o m en k an k an n a y i zh ong q iu j i b i j iao h ao . _|0 4 4 3 3 5 5 4 4 5 5 3 3 4 4 3 3 2 2 4 4 3 3 4 4 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000651|0001_Angry|ZH|播放歌单收藏,脑筋可动得真快.|_ b o f ang g e d an sh ou c ang , n ao j in k e d ong d e zh en k uai . _|0 1 1 4 4 1 1 1 1 1 1 2 2 0 3 3 1 1 3 3 4 4 5 5 1 1 4 4 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000647|0001_Angry|ZH|三个,两个儿子一个女儿.|_ s an g e , l iang g e EE er z i0 y i g e n v EE er . _|0 1 1 5 5 0 3 3 5 5 2 2 5 5 2 2 5 5 3 3 2 2 0 0|1 2 2 1 2 2 2 2 2 2 2 2 1 1
+0001_000445|0001_Angry|ZH|我倒是有一个爱好收藏古董.|_ w o d ao sh ir y ou y i g e AA ai h ao sh ou c ang g u d ong . _|0 3 3 4 4 4 4 3 3 2 2 5 5 4 4 4 4 1 1 2 2 2 2 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000366|0001_Angry|ZH|他一定是一眼就被你迷住了.|_ t a y i d ing sh ir y i y En j iu b ei n i m i zh u l e . _|0 1 1 2 2 4 4 4 4 4 4 3 3 4 4 4 4 3 3 2 2 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000425|0001_Angry|ZH|我也想去看可爱的熊猫.|_ w o y E x iang q v k an k e AA ai d e x iong m ao . _|0 3 3 2 2 3 3 4 4 4 4 3 3 4 4 5 5 2 2 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000481|0001_Angry|ZH|吝啬鬼!他每天还骑自行车上学!|_ l in s e g ui ! t a m ei t ian h ai q i z i0 x ing ch e sh ang x ve ! _|0 4 4 4 4 3 3 0 1 1 3 3 1 1 2 2 2 2 4 4 2 2 1 1 4 4 2 2 0 0|1 2 2 2 1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000487|0001_Angry|ZH|也许能帮助你把事情弄清楚.|_ y E x v n eng b ang zh u n i b a sh ir q ing n ong q ing ch u . _|0 2 2 3 3 2 2 1 1 4 4 2 2 3 3 4 4 5 5 4 4 1 1 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000532|0001_Angry|ZH|大概足够支持我生活三个月的.|_ d a g ai z u g ou zh ir ch ir w o sh eng h uo s an g e y ve d e . _|0 4 4 4 4 2 2 4 4 1 1 2 2 3 3 1 1 2 2 1 1 5 5 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000694|0001_Angry|ZH|很漂亮,不过人多拥挤.|_ h en p iao l iang , b u g uo r en d uo y ong j i . _|0 3 3 4 4 5 5 0 2 2 4 4 2 2 1 1 1 1 3 3 0 0|1 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000545|0001_Angry|ZH|我以为你们国家的人都是麻将高手.|_ w o y i w ei n i m en g uo j ia d e r en d ou sh ir m a j iang g ao sh ou . _|0 2 2 3 3 2 2 3 3 5 5 2 2 1 1 5 5 2 2 1 1 4 4 2 2 1 1 1 1 3 3 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000437|0001_Angry|ZH|是没什么但是挺别扭的.|_ sh ir m ei sh en m e d an sh ir t ing b ie n iu d e . _|0 4 4 2 2 2 2 5 5 4 4 4 4 3 3 4 4 5 5 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000531|0001_Angry|ZH|时间对珍尼来说是没有用的.|_ sh ir j ian d ui zh en n i l ai sh uo sh ir m ei y ou y ong d e . _|0 2 2 1 1 4 4 1 1 2 2 2 2 1 1 4 4 2 2 3 3 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000594|0001_Angry|ZH|我的直系亲属人数不多.|_ w o d e zh ir x i q in sh u r en sh u b u d uo . _|0 3 3 5 5 2 2 4 4 1 1 3 3 2 2 4 4 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000539|0001_Angry|ZH|我是一名教师,你可是好眼光.|_ w o sh ir y i m ing j iao sh ir , n i k e sh ir h ao y En g uang . _|0 3 3 4 4 4 4 2 2 4 4 1 1 0 2 2 3 3 4 4 2 2 3 3 1 1 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000497|0001_Angry|ZH|不知道.或许一双新鞋.|_ b u zh ir d ao . h uo x v y i sh uang x in x ie . _|0 4 4 1 1 4 4 0 4 4 3 3 4 4 1 1 1 1 2 2 0 0|1 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000452|0001_Angry|ZH|我昨天遇到马克,他看起来很忧郁.|_ w o z uo t ian y v d ao m a k e , t a k an q i l ai h en y ou y v . _|0 3 3 2 2 1 1 4 4 4 4 3 3 4 4 0 1 1 4 4 3 3 5 5 3 3 1 1 4 4 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000684|0001_Angry|ZH|我曾经养过,我太高兴了.|_ w o c eng j ing y ang g uo , w o t ai g ao x ing l e . _|0 3 3 2 2 1 1 3 3 4 4 0 3 3 4 4 1 1 4 4 5 5 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000625|0001_Angry|ZH|然后再找一个音乐播放器,|_ r an h ou z ai zh ao y i g e y in y ve b o f ang q i , _|0 2 2 4 4 4 4 3 3 2 2 5 5 1 1 4 4 1 1 4 4 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000502|0001_Angry|ZH|小心脚下.人行道上有个坑.|_ x iao x in j iao x ia . r en x ing d ao sh ang y ou g e k eng . _|0 3 3 1 1 3 3 5 5 0 2 2 2 2 4 4 4 4 3 3 5 5 1 1 0 0|1 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000419|0001_Angry|ZH|以后不要喝那么多了,伤身体.|_ y i h ou b u y ao h e n a m e d uo l e , sh ang sh en t i . _|0 3 3 4 4 2 2 4 4 1 1 4 4 5 5 1 1 5 5 0 1 1 1 1 3 3 0 0|1 2 2 2 2 2 2 2 2 2 1 2 2 2 1 1
+0001_000483|0001_Angry|ZH|是不是依然觉得我很可爱.|_ sh ir b u sh ir y i r an j ve d e w o h en k e AA ai . _|0 4 4 5 5 4 4 1 1 2 2 2 2 5 5 2 2 3 3 3 3 4 4 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000364|0001_Angry|ZH|谁你也不认识,我很乐意帮助你.|_ sh ui n i y E b u r en sh ir , w o h en l e y i b ang zh u n i . _|0 2 2 2 2 3 3 2 2 4 4 5 5 0 2 2 3 3 4 4 4 4 1 1 4 4 3 3 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000620|0001_Angry|ZH|没有找到你想删除的闹钟.|_ m ei y ou zh ao d ao n i x iang sh an ch u d e n ao zh ong . _|0 2 2 3 3 3 3 4 4 2 2 3 3 1 1 2 2 5 5 4 4 1 1 0 0|1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000687|0001_Angry|ZH|你看起来很高兴,眼睛闪闪发亮.|_ n i k an q i l ai h en g ao x ing , y En j ing sh an sh an f a l iang . _|0 3 3 4 4 3 3 5 5 3 3 1 1 4 4 0 3 3 5 5 3 3 5 5 1 1 4 4 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 2 1 1
+0001_000678|0001_Angry|ZH|太棒了,我们下午可以在湖里划船.|_ t ai b ang l e , w o m en x ia w u k e y i z ai h u l i h ua ch uan . _|0 4 4 4 4 5 5 0 3 3 5 5 4 4 3 3 2 2 3 3 4 4 2 2 5 5 2 2 2 2 0 0|1 2 2 2 1 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000444|0001_Angry|ZH|很神奇的样子,我搞不懂为什么.|_ h en sh en q i d e y ang z i0 , w o g ao b u d ong w ei sh en m e . _|0 3 3 2 2 2 2 5 5 4 4 5 5 0 3 3 3 3 5 5 3 3 4 4 2 2 5 5 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000494|0001_Angry|ZH|节食减肥很痛苦.|_ j ie sh ir j ian f ei h en t ong k u . _|0 2 2 2 2 3 3 2 2 3 3 4 4 3 3 0 0|1 2 2 2 2 2 2 2 1 1
+0001_000695|0001_Angry|ZH|别总是闲着,找点事情干.|_ b ie z ong sh ir x ian zh e , zh ao d ian sh ir q ing g an . _|0 2 2 3 3 4 4 2 2 5 5 0 2 2 3 3 4 4 5 5 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000546|0001_Angry|ZH|我饿啦,我想去吃点东西.|_ w o EE e l a , w o x iang q v ch ir d ian d ong x i . _|0 3 3 4 4 5 5 0 2 2 3 3 4 4 1 1 3 3 1 1 5 5 0 0|1 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000637|0001_Angry|ZH|我太喜欢听了,所以不断重复着听.|_ w o t ai x i h uan t ing l e , s uo y i b u d uan ch ong f u zh e t ing . _|0 3 3 4 4 3 3 5 5 1 1 5 5 0 2 2 3 3 2 2 4 4 2 2 4 4 5 5 1 1 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 2 2 2 1 1
+0001_000474|0001_Angry|ZH|是的,请返还我的钱,谢谢.|_ sh ir d e , q ing f an h uan w o d e q ian , x ie x ie . _|0 4 4 5 5 0 2 2 3 3 2 2 3 3 5 5 2 2 0 4 4 5 5 0 0|1 2 2 1 2 2 2 2 2 2 1 2 2 1 1
+0001_000697|0001_Angry|ZH|这两块是唐朝不同时期铸造的.|_ zh e l iang k uai sh ir t ang ch ao b u t ong sh ir q i zh u z ao d e . _|0 4 4 3 3 4 4 4 4 2 2 2 2 4 4 2 2 2 2 1 1 4 4 4 4 5 5 0 0|1 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1
+0001_000591|0001_Angry|ZH|请勿进入竹林.不让进.|_ q ing w u j in r u zh u l in . b u r ang j in . _|0 3 3 4 4 4 4 4 4 2 2 2 2 0 2 2 4 4 4 4 0 0|1 2 2 2 2 2 2 1 2 2 2 1 1
+0001_000455|0001_Angry|ZH|我们俩合不来,还经常吵架.|_ w o m en l ia h e b u l ai , h ai j ing ch ang ch ao j ia . _|0 3 3 5 5 3 3 2 2 5 5 2 2 0 2 2 1 1 2 2 3 3 4 4 0 0|1 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000649|0001_Angry|ZH|是的,我刚撞到了桌子.|_ sh ir d e , w o g ang zh uang d ao l e zh uo z i0 . _|0 4 4 5 5 0 3 3 1 1 4 4 4 4 5 5 1 1 5 5 0 0|1 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000442|0001_Angry|ZH|不会这么凑巧吧!我也是十六.|_ b u h ui zh e m e c ou q iao b a ! w o y E sh ir sh ir l iu . _|0 2 2 4 4 4 4 5 5 4 4 3 3 5 5 0 2 2 3 3 4 4 2 2 4 4 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000552|0001_Angry|ZH|那就一会再说,我好害怕.|_ n a j iu y i h ui z ai sh uo , w o h ao h ai p a . _|0 4 4 4 4 2 2 4 4 4 4 1 1 0 2 2 3 3 4 4 4 4 0 0|1 2 2 2 2 2 2 1 2 2 2 2 1 1
+0001_000361|0001_Angry|ZH|妇女节快乐.我永远爱你,妈妈.|_ f u n v j ie k uai l e . w o y ong y van AA ai n i , m a m a . _|0 4 4 3 3 2 2 4 4 4 4 0 3 3 2 2 3 3 4 4 3 3 0 1 1 5 5 0 0|1 2 2 2 2 2 1 2 2 2 2 2 1 2 2 1 1
+0001_000426|0001_Angry|ZH|是的,你是个大块头,我是守门员.|_ sh ir d e , n i sh ir g e d a k uai t ou , w o sh ir sh ou m en y van . _|0 4 4 5 5 0 3 3 4 4 5 5 4 4 4 4 2 2 0 3 3 4 4 3 3 2 2 2 2 0 0|1 2 2 1 2 2 2 2 2 2 1 2 2 2 2 2 1 1
+0001_000519|0001_Angry|ZH|让我再想想,真相已经上传了.|_ r ang w o z ai x iang x iang , zh en x iang y i j ing sh ang ch uan l e . _|0 4 4 3 3 4 4 3 3 5 5 0 1 1 4 4 3 3 1 1 4 4 2 2 5 5 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1

filelists/val.list

@@ -0,0 +1,2 @@
+0001_000529|0001_Angry|ZH|她和维克分手了,所以她申请转调.|_ t a h e w ei k e f en sh ou l e , s uo y i t a sh en q ing zh uan d iao . _|0 1 1 2 2 2 2 4 4 1 1 3 3 5 5 0 2 2 3 3 1 1 1 1 3 3 3 3 4 4 0 0|1 2 2 2 2 2 2 2 1 2 2 2 2 2 2 2 1 1
+0001_000443|0001_Angry|ZH|别不好意思.再多吃些鸡肉.|_ b ie b u h ao y i s i0 . z ai d uo ch ir x ie j i r ou . _|0 2 2 4 4 3 3 4 4 5 5 0 4 4 1 1 1 1 1 1 1 1 4 4 0 0|1 2 2 2 2 2 1 2 2 2 2 2 2 1 1

losses.py

@@ -0,0 +1,61 @@
+import torch 
+from torch.nn import functional as F
+
+import commons
+
+
+def feature_loss(fmap_r, fmap_g):
+  loss = 0
+  for dr, dg in zip(fmap_r, fmap_g):
+    for rl, gl in zip(dr, dg):
+      rl = rl.float().detach()
+      gl = gl.float()
+      loss += torch.mean(torch.abs(rl - gl))
+
+  return loss * 2 
+
+
+def discriminator_loss(disc_real_outputs, disc_generated_outputs):
+  loss = 0
+  r_losses = []
+  g_losses = []
+  for dr, dg in zip(disc_real_outputs, disc_generated_outputs):
+    dr = dr.float()
+    dg = dg.float()
+    r_loss = torch.mean((1-dr)**2)
+    g_loss = torch.mean(dg**2)
+    loss += (r_loss + g_loss)
+    r_losses.append(r_loss.item())
+    g_losses.append(g_loss.item())
+
+  return loss, r_losses, g_losses
+
+
+def generator_loss(disc_outputs):
+  loss = 0
+  gen_losses = []
+  for dg in disc_outputs:
+    dg = dg.float()
+    l = torch.mean((1-dg)**2)
+    gen_losses.append(l)
+    loss += l
+
+  return loss, gen_losses
+
+
+def kl_loss(z_p, logs_q, m_p, logs_p, z_mask):
+  """
+  z_p, logs_q: [b, h, t_t]
+  m_p, logs_p: [b, h, t_t]
+  """
+  z_p = z_p.float()
+  logs_q = logs_q.float()
+  m_p = m_p.float()
+  logs_p = logs_p.float()
+  z_mask = z_mask.float()
+
+  kl = logs_p - logs_q - 0.5
+  kl += 0.5 * ((z_p - m_p)**2) * torch.exp(-2. * logs_p)
+  kl = torch.sum(kl * z_mask)
+  l = kl / torch.sum(z_mask)
+  return l

mel_processing.py

@@ -0,0 +1,112 @@
+import math
+import os
+import random
+import torch
+from torch import nn
+import torch.nn.functional as F
+import torch.utils.data
+import numpy as np
+import librosa
+import librosa.util as librosa_util
+from librosa.util import normalize, pad_center, tiny
+from scipy.signal import get_window
+from scipy.io.wavfile import read
+from librosa.filters import mel as librosa_mel_fn
+
+MAX_WAV_VALUE = 32768.0
+
+
+def dynamic_range_compression_torch(x, C=1, clip_val=1e-5):
+    """
+    PARAMS
+    ------
+    C: compression factor
+    """
+    return torch.log(torch.clamp(x, min=clip_val) * C)
+
+
+def dynamic_range_decompression_torch(x, C=1):
+    """
+    PARAMS
+    ------
+    C: compression factor used to compress
+    """
+    return torch.exp(x) / C
+
+
+def spectral_normalize_torch(magnitudes):
+    output = dynamic_range_compression_torch(magnitudes)
+    return output
+
+
+def spectral_de_normalize_torch(magnitudes):
+    output = dynamic_range_decompression_torch(magnitudes)
+    return output
+
+
+mel_basis = {}
+hann_window = {}
+
+
+def spectrogram_torch(y, n_fft, sampling_rate, hop_size, win_size, center=False):
+    if torch.min(y) < -1.:
+        print('min value is ', torch.min(y))
+    if torch.max(y) > 1.:
+        print('max value is ', torch.max(y))
+
+    global hann_window
+    dtype_device = str(y.dtype) + '_' + str(y.device)
+    wnsize_dtype_device = str(win_size) + '_' + dtype_device
+    if wnsize_dtype_device not in hann_window:
+        hann_window[wnsize_dtype_device] = torch.hann_window(win_size).to(dtype=y.dtype, device=y.device)
+
+    y = torch.nn.functional.pad(y.unsqueeze(1), (int((n_fft-hop_size)/2), int((n_fft-hop_size)/2)), mode='reflect')
+    y = y.squeeze(1)
+
+    spec = torch.stft(y, n_fft, hop_length=hop_size, win_length=win_size, window=hann_window[wnsize_dtype_device],
+                      center=center, pad_mode='reflect', normalized=False, onesided=True, return_complex=False)
+
+    spec = torch.sqrt(spec.pow(2).sum(-1) + 1e-6)
+    return spec
+
+
+def spec_to_mel_torch(spec, n_fft, num_mels, sampling_rate, fmin, fmax):
+    global mel_basis
+    dtype_device = str(spec.dtype) + '_' + str(spec.device)
+    fmax_dtype_device = str(fmax) + '_' + dtype_device
+    if fmax_dtype_device not in mel_basis:
+        mel = librosa_mel_fn(sampling_rate, n_fft, num_mels, fmin, fmax)
+        mel_basis[fmax_dtype_device] = torch.from_numpy(mel).to(dtype=spec.dtype, device=spec.device)
+    spec = torch.matmul(mel_basis[fmax_dtype_device], spec)
+    spec = spectral_normalize_torch(spec)
+    return spec
+
+
+def mel_spectrogram_torch(y, n_fft, num_mels, sampling_rate, hop_size, win_size, fmin, fmax, center=False):
+    if torch.min(y) < -1.:
+        print('min value is ', torch.min(y))
+    if torch.max(y) > 1.:
+        print('max value is ', torch.max(y))
+
+    global mel_basis, hann_window
+    dtype_device = str(y.dtype) + '_' + str(y.device)
+    fmax_dtype_device = str(fmax) + '_' + dtype_device
+    wnsize_dtype_device = str(win_size) + '_' + dtype_device
+    if fmax_dtype_device not in mel_basis:
+        mel = librosa_mel_fn(sampling_rate, n_fft, num_mels, fmin, fmax)
+        mel_basis[fmax_dtype_device] = torch.from_numpy(mel).to(dtype=y.dtype, device=y.device)
+    if wnsize_dtype_device not in hann_window:
+        hann_window[wnsize_dtype_device] = torch.hann_window(win_size).to(dtype=y.dtype, device=y.device)
+
+    y = torch.nn.functional.pad(y.unsqueeze(1), (int((n_fft-hop_size)/2), int((n_fft-hop_size)/2)), mode='reflect')
+    y = y.squeeze(1)
+
+    spec = torch.stft(y, n_fft, hop_length=hop_size, win_length=win_size, window=hann_window[wnsize_dtype_device],
+                      center=center, pad_mode='reflect', normalized=False, onesided=True, return_complex=False)
+
+    spec = torch.sqrt(spec.pow(2).sum(-1) + 1e-6)
+
+    spec = torch.matmul(mel_basis[fmax_dtype_device], spec)
+    spec = spectral_normalize_torch(spec)
+
+    return spec

models.py

@@ -0,0 +1,596 @@
+import copy
+import math
+import torch
+from torch import nn
+from torch.nn import functional as F
+
+import commons
+import modules
+import attentions
+import monotonic_align
+
+from torch.nn import Conv1d, ConvTranspose1d, AvgPool1d, Conv2d
+from torch.nn.utils import weight_norm, remove_weight_norm, spectral_norm
+
+from GST import GST
+from commons import init_weights, get_padding
+from text import symbols, num_tones, num_languages
+
+
+class StochasticDurationPredictor(nn.Module):
+    def __init__(self, in_channels, filter_channels, kernel_size, p_dropout, n_flows=4, gin_channels=0):
+        super().__init__()
+        filter_channels = in_channels  # it needs to be removed from future version.
+        self.in_channels = in_channels
+        self.filter_channels = filter_channels
+        self.kernel_size = kernel_size
+        self.p_dropout = p_dropout
+        self.n_flows = n_flows
+        self.gin_channels = gin_channels
+
+        self.log_flow = modules.Log()
+        self.flows = nn.ModuleList()
+        self.flows.append(modules.ElementwiseAffine(2))
+        for i in range(n_flows):
+            self.flows.append(modules.ConvFlow(2, filter_channels, kernel_size, n_layers=3))
+            self.flows.append(modules.Flip())
+
+        self.post_pre = nn.Conv1d(1, filter_channels, 1)
+        self.post_proj = nn.Conv1d(filter_channels, filter_channels, 1)
+        self.post_convs = modules.DDSConv(filter_channels, kernel_size, n_layers=3, p_dropout=p_dropout)
+        self.post_flows = nn.ModuleList()
+        self.post_flows.append(modules.ElementwiseAffine(2))
+        for i in range(4):
+            self.post_flows.append(modules.ConvFlow(2, filter_channels, kernel_size, n_layers=3))
+            self.post_flows.append(modules.Flip())
+
+        self.pre = nn.Conv1d(in_channels, filter_channels, 1)
+        self.proj = nn.Conv1d(filter_channels, filter_channels, 1)
+        self.convs = modules.DDSConv(filter_channels, kernel_size, n_layers=3, p_dropout=p_dropout)
+        if gin_channels != 0:
+            self.cond = nn.Conv1d(gin_channels, filter_channels, 1)
+
+    def forward(self, x, x_mask, w=None, g=None, reverse=False, noise_scale=1.0):
+        x = torch.detach(x)
+        x = self.pre(x)
+        if g is not None:
+            g = torch.detach(g)
+            x = x + self.cond(g)
+        x = self.convs(x, x_mask)
+        x = self.proj(x) * x_mask
+
+        if not reverse:
+            flows = self.flows
+            assert w is not None
+
+            logdet_tot_q = 0
+            h_w = self.post_pre(w)
+            h_w = self.post_convs(h_w, x_mask)
+            h_w = self.post_proj(h_w) * x_mask
+            e_q = torch.randn(w.size(0), 2, w.size(2)).to(device=x.device, dtype=x.dtype) * x_mask
+            z_q = e_q
+            for flow in self.post_flows:
+                z_q, logdet_q = flow(z_q, x_mask, g=(x + h_w))
+                logdet_tot_q += logdet_q
+            z_u, z1 = torch.split(z_q, [1, 1], 1)
+            u = torch.sigmoid(z_u) * x_mask
+            z0 = (w - u) * x_mask
+            logdet_tot_q += torch.sum((F.logsigmoid(z_u) + F.logsigmoid(-z_u)) * x_mask, [1, 2])
+            logq = torch.sum(-0.5 * (math.log(2 * math.pi) + (e_q ** 2)) * x_mask, [1, 2]) - logdet_tot_q
+
+            logdet_tot = 0
+            z0, logdet = self.log_flow(z0, x_mask)
+            logdet_tot += logdet
+            z = torch.cat([z0, z1], 1)
+            for flow in flows:
+                z, logdet = flow(z, x_mask, g=x, reverse=reverse)
+                logdet_tot = logdet_tot + logdet
+            nll = torch.sum(0.5 * (math.log(2 * math.pi) + (z ** 2)) * x_mask, [1, 2]) - logdet_tot
+            return nll + logq  # [b]
+        else:
+            flows = list(reversed(self.flows))
+            flows = flows[:-2] + [flows[-1]]  # remove a useless vflow
+            z = torch.randn(x.size(0), 2, x.size(2)).to(device=x.device, dtype=x.dtype) * noise_scale
+            for flow in flows:
+                z = flow(z, x_mask, g=x, reverse=reverse)
+            z0, z1 = torch.split(z, [1, 1], 1)
+            logw = z0
+            return logw
+
+
+class DurationPredictor(nn.Module):
+    def __init__(self, in_channels, filter_channels, kernel_size, p_dropout, gin_channels=0):
+        super().__init__()
+
+        self.in_channels = in_channels
+        self.filter_channels = filter_channels
+        self.kernel_size = kernel_size
+        self.p_dropout = p_dropout
+        self.gin_channels = gin_channels
+
+        self.drop = nn.Dropout(p_dropout)
+        self.conv_1 = nn.Conv1d(in_channels, filter_channels, kernel_size, padding=kernel_size // 2)
+        self.norm_1 = modules.LayerNorm(filter_channels)
+        self.conv_2 = nn.Conv1d(filter_channels, filter_channels, kernel_size, padding=kernel_size // 2)
+        self.norm_2 = modules.LayerNorm(filter_channels)
+        self.proj = nn.Conv1d(filter_channels, 1, 1)
+
+        if gin_channels != 0:
+            self.cond = nn.Conv1d(gin_channels, in_channels, 1)
+
+    def forward(self, x, x_mask, g=None):
+        x = torch.detach(x)
+        if g is not None:
+            g = torch.detach(g)
+            x = x + self.cond(g)
+        x = self.conv_1(x * x_mask)
+        x = torch.relu(x)
+        x = self.norm_1(x)
+        x = self.drop(x)
+        x = self.conv_2(x * x_mask)
+        x = torch.relu(x)
+        x = self.norm_2(x)
+        x = self.drop(x)
+        x = self.proj(x * x_mask)
+        return x * x_mask
+
+
+class TextEncoder(nn.Module):
+    def __init__(self,
+                 n_vocab,
+                 out_channels,
+                 hidden_channels,
+                 filter_channels,
+                 n_heads,
+                 n_layers,
+                 kernel_size,
+                 p_dropout):
+        super().__init__()
+        self.n_vocab = n_vocab
+        self.out_channels = out_channels
+        self.hidden_channels = hidden_channels
+        self.filter_channels = filter_channels
+        self.n_heads = n_heads
+        self.n_layers = n_layers
+        self.kernel_size = kernel_size
+        self.p_dropout = p_dropout
+
+        self.emb = nn.Embedding(len(symbols), hidden_channels)
+        nn.init.normal_(self.emb.weight, 0.0, hidden_channels ** -0.5)
+        self.tone_emb = nn.Embedding(num_tones, hidden_channels)
+        nn.init.normal_(self.tone_emb.weight, 0.0, hidden_channels ** -0.5)
+        self.language_emb = nn.Embedding(num_languages, hidden_channels)
+        nn.init.normal_(self.language_emb.weight, 0.0, hidden_channels ** -0.5)
+        self.bert_proj = nn.Conv1d(1024, hidden_channels, 1)
+
+        self.encoder = attentions.Encoder(
+            hidden_channels,
+            filter_channels,
+            n_heads,
+            n_layers,
+            kernel_size,
+            p_dropout)
+        self.proj = nn.Conv1d(hidden_channels, out_channels * 2, 1)
+
+    def forward(self, x, x_lengths, tone, language, bert):
+        x = (self.emb(x)+ self.tone_emb(tone)+ self.language_emb(language) +self.bert_proj(bert)) * math.sqrt(self.hidden_channels)  # [b, t, h]
+        x = torch.transpose(x, 1, -1)  # [b, h, t]
+        x_mask = torch.unsqueeze(commons.sequence_mask(x_lengths, x.size(2)), 1).to(x.dtype)
+
+        x = self.encoder(x * x_mask, x_mask)
+        stats = self.proj(x) * x_mask
+
+        m, logs = torch.split(stats, self.out_channels, dim=1)
+        return x, m, logs, x_mask
+
+
+class ResidualCouplingBlock(nn.Module):
+    def __init__(self,
+                 channels,
+                 hidden_channels,
+                 kernel_size,
+                 dilation_rate,
+                 n_layers,
+                 n_flows=4,
+                 gin_channels=0):
+        super().__init__()
+        self.channels = channels
+        self.hidden_channels = hidden_channels
+        self.kernel_size = kernel_size
+        self.dilation_rate = dilation_rate
+        self.n_layers = n_layers
+        self.n_flows = n_flows
+        self.gin_channels = gin_channels
+
+        self.flows = nn.ModuleList()
+        for i in range(n_flows):
+            self.flows.append(
+                modules.ResidualCouplingLayer(channels, hidden_channels, kernel_size, dilation_rate, n_layers,
+                                              gin_channels=gin_channels, mean_only=True))
+            self.flows.append(modules.Flip())
+
+    def forward(self, x, x_mask, g=None, reverse=False):
+        if not reverse:
+            for flow in self.flows:
+                x, _ = flow(x, x_mask, g=g, reverse=reverse)
+        else:
+            for flow in reversed(self.flows):
+                x = flow(x, x_mask, g=g, reverse=reverse)
+        return x
+
+
+class PosteriorEncoder(nn.Module):
+    def __init__(self,
+                 in_channels,
+                 out_channels,
+                 hidden_channels,
+                 kernel_size,
+                 dilation_rate,
+                 n_layers,
+                 gin_channels=0):
+        super().__init__()
+        self.in_channels = in_channels
+        self.out_channels = out_channels
+        self.hidden_channels = hidden_channels
+        self.kernel_size = kernel_size
+        self.dilation_rate = dilation_rate
+        self.n_layers = n_layers
+        self.gin_channels = gin_channels
+
+        self.pre = nn.Conv1d(in_channels, hidden_channels, 1)
+        self.enc = modules.WN(hidden_channels, kernel_size, dilation_rate, n_layers, gin_channels=gin_channels)
+        self.proj = nn.Conv1d(hidden_channels, out_channels * 2, 1)
+
+    def forward(self, x, x_lengths, g=None):
+        x_mask = torch.unsqueeze(commons.sequence_mask(x_lengths, x.size(2)), 1).to(x.dtype)
+        x = self.pre(x) * x_mask
+        x = self.enc(x, x_mask, g=g)
+        stats = self.proj(x) * x_mask
+        m, logs = torch.split(stats, self.out_channels, dim=1)
+        z = (m + torch.randn_like(m) * torch.exp(logs)) * x_mask
+        return z, m, logs, x_mask
+
+
+class Generator(torch.nn.Module):
+    def __init__(self, initial_channel, resblock, resblock_kernel_sizes, resblock_dilation_sizes, upsample_rates,
+                 upsample_initial_channel, upsample_kernel_sizes, gin_channels=0):
+        super(Generator, self).__init__()
+        self.num_kernels = len(resblock_kernel_sizes)
+        self.num_upsamples = len(upsample_rates)
+        self.conv_pre = Conv1d(initial_channel, upsample_initial_channel, 7, 1, padding=3)
+        resblock = modules.ResBlock1 if resblock == '1' else modules.ResBlock2
+
+        self.ups = nn.ModuleList()
+        for i, (u, k) in enumerate(zip(upsample_rates, upsample_kernel_sizes)):
+            self.ups.append(weight_norm(
+                ConvTranspose1d(upsample_initial_channel // (2 ** i), upsample_initial_channel // (2 ** (i + 1)),
+                                k, u, padding=(k - u) // 2)))
+
+        self.resblocks = nn.ModuleList()
+        for i in range(len(self.ups)):
+            ch = upsample_initial_channel // (2 ** (i + 1))
+            for j, (k, d) in enumerate(zip(resblock_kernel_sizes, resblock_dilation_sizes)):
+                self.resblocks.append(resblock(ch, k, d))
+
+        self.conv_post = Conv1d(ch, 1, 7, 1, padding=3, bias=False)
+        self.ups.apply(init_weights)
+
+        if gin_channels != 0:
+            self.cond = nn.Conv1d(gin_channels, upsample_initial_channel, 1)
+
+    def forward(self, x, g=None):
+        x = self.conv_pre(x)
+        if g is not None:
+            x = x + self.cond(g)
+
+        for i in range(self.num_upsamples):
+            x = F.leaky_relu(x, modules.LRELU_SLOPE)
+            x = self.ups[i](x)
+            xs = None
+            for j in range(self.num_kernels):
+                if xs is None:
+                    xs = self.resblocks[i * self.num_kernels + j](x)
+                else:
+                    xs += self.resblocks[i * self.num_kernels + j](x)
+            x = xs / self.num_kernels
+        x = F.leaky_relu(x)
+        x = self.conv_post(x)
+        x = torch.tanh(x)
+
+        return x
+
+    def remove_weight_norm(self):
+        print('Removing weight norm...')
+        for l in self.ups:
+            remove_weight_norm(l)
+        for l in self.resblocks:
+            l.remove_weight_norm()
+
+
+class DiscriminatorP(torch.nn.Module):
+    def __init__(self, period, kernel_size=5, stride=3, use_spectral_norm=False):
+        super(DiscriminatorP, self).__init__()
+        self.period = period
+        self.use_spectral_norm = use_spectral_norm
+        norm_f = weight_norm if use_spectral_norm == False else spectral_norm
+        self.convs = nn.ModuleList([
+            norm_f(Conv2d(1, 32, (kernel_size, 1), (stride, 1), padding=(get_padding(kernel_size, 1), 0))),
+            norm_f(Conv2d(32, 128, (kernel_size, 1), (stride, 1), padding=(get_padding(kernel_size, 1), 0))),
+            norm_f(Conv2d(128, 512, (kernel_size, 1), (stride, 1), padding=(get_padding(kernel_size, 1), 0))),
+            norm_f(Conv2d(512, 1024, (kernel_size, 1), (stride, 1), padding=(get_padding(kernel_size, 1), 0))),
+            norm_f(Conv2d(1024, 1024, (kernel_size, 1), 1, padding=(get_padding(kernel_size, 1), 0))),
+        ])
+        self.conv_post = norm_f(Conv2d(1024, 1, (3, 1), 1, padding=(1, 0)))
+
+    def forward(self, x):
+        fmap = []
+
+        # 1d to 2d
+        b, c, t = x.shape
+        if t % self.period != 0:  # pad first
+            n_pad = self.period - (t % self.period)
+            x = F.pad(x, (0, n_pad), "reflect")
+            t = t + n_pad
+        x = x.view(b, c, t // self.period, self.period)
+
+        for l in self.convs:
+            x = l(x)
+            x = F.leaky_relu(x, modules.LRELU_SLOPE)
+            fmap.append(x)
+        x = self.conv_post(x)
+        fmap.append(x)
+        x = torch.flatten(x, 1, -1)
+
+        return x, fmap
+
+
+class DiscriminatorS(torch.nn.Module):
+    def __init__(self, use_spectral_norm=False):
+        super(DiscriminatorS, self).__init__()
+        norm_f = weight_norm if use_spectral_norm == False else spectral_norm
+        self.convs = nn.ModuleList([
+            norm_f(Conv1d(1, 16, 15, 1, padding=7)),
+            norm_f(Conv1d(16, 64, 41, 4, groups=4, padding=20)),
+            norm_f(Conv1d(64, 256, 41, 4, groups=16, padding=20)),
+            norm_f(Conv1d(256, 1024, 41, 4, groups=64, padding=20)),
+            norm_f(Conv1d(1024, 1024, 41, 4, groups=256, padding=20)),
+            norm_f(Conv1d(1024, 1024, 5, 1, padding=2)),
+        ])
+        self.conv_post = norm_f(Conv1d(1024, 1, 3, 1, padding=1))
+
+    def forward(self, x):
+        fmap = []
+
+        for l in self.convs:
+            x = l(x)
+            x = F.leaky_relu(x, modules.LRELU_SLOPE)
+            fmap.append(x)
+        x = self.conv_post(x)
+        fmap.append(x)
+        x = torch.flatten(x, 1, -1)
+
+        return x, fmap
+
+
+class MultiPeriodDiscriminator(torch.nn.Module):
+    def __init__(self, use_spectral_norm=False):
+        super(MultiPeriodDiscriminator, self).__init__()
+        periods = [2, 3, 5, 7, 11]
+
+        discs = [DiscriminatorS(use_spectral_norm=use_spectral_norm)]
+        discs = discs + [DiscriminatorP(i, use_spectral_norm=use_spectral_norm) for i in periods]
+        self.discriminators = nn.ModuleList(discs)
+
+    def forward(self, y, y_hat):
+        y_d_rs = []
+        y_d_gs = []
+        fmap_rs = []
+        fmap_gs = []
+        for i, d in enumerate(self.discriminators):
+            y_d_r, fmap_r = d(y)
+            y_d_g, fmap_g = d(y_hat)
+            y_d_rs.append(y_d_r)
+            y_d_gs.append(y_d_g)
+            fmap_rs.append(fmap_r)
+            fmap_gs.append(fmap_g)
+
+        return y_d_rs, y_d_gs, fmap_rs, fmap_gs
+
+class ReferenceEncoder(nn.Module):
+    '''
+    inputs --- [N, Ty/r, n_mels*r]  mels
+    outputs --- [N, ref_enc_gru_size]
+    '''
+
+    def __init__(self, spec_channels):
+
+        super().__init__()
+        self.spec_channels = spec_channels
+        ref_enc_filters = [32, 32, 64, 64, 128, 128]
+        K = len(ref_enc_filters)
+        filters = [1] + ref_enc_filters
+        convs = [weight_norm(nn.Conv2d(in_channels=filters[i],
+                                       out_channels=filters[i + 1],
+                                       kernel_size=(3, 3),
+                                       stride=(2, 2),
+                                       padding=(1, 1))) for i in range(K)]
+        self.convs = nn.ModuleList(convs)
+        # self.wns = nn.ModuleList([weight_norm(num_features=ref_enc_filters[i]) for i in range(K)])
+
+        out_channels = self.calculate_channels(spec_channels, 3, 2, 1, K)
+        self.gru = nn.GRU(input_size=ref_enc_filters[-1] * out_channels,
+                          hidden_size=256 // 2,
+                          batch_first=True)
+
+    def forward(self, inputs, mask=None):
+        N = inputs.size(0)
+        out = inputs.view(N, 1, -1, self.spec_channels)  # [N, 1, Ty, n_freqs]
+        for conv in self.convs:
+            out = conv(out)
+            # out = wn(out)
+            out = F.relu(out)  # [N, 128, Ty//2^K, n_mels//2^K]
+
+        out = out.transpose(1, 2)  # [N, Ty//2^K, 128, n_mels//2^K]
+        T = out.size(1)
+        N = out.size(0)
+        out = out.contiguous().view(N, T, -1)  # [N, Ty//2^K, 128*n_mels//2^K]
+
+        self.gru.flatten_parameters()
+        memory, out = self.gru(out)  # out --- [1, N, 128]
+
+        return out.squeeze(0)
+
+    def calculate_channels(self, L, kernel_size, stride, pad, n_convs):
+        for i in range(n_convs):
+            L = (L - kernel_size + 2 * pad) // stride + 1
+        return L
+
+
+class SynthesizerTrn(nn.Module):
+    """
+  Synthesizer for Training
+  """
+
+    def __init__(self,
+                 n_vocab,
+                 spec_channels,
+                 segment_size,
+                 inter_channels,
+                 hidden_channels,
+                 filter_channels,
+                 n_heads,
+                 n_layers,
+                 kernel_size,
+                 p_dropout,
+                 resblock,
+                 resblock_kernel_sizes,
+                 resblock_dilation_sizes,
+                 upsample_rates,
+                 upsample_initial_channel,
+                 upsample_kernel_sizes,
+                 n_speakers=0,
+                 gin_channels=0,
+                 use_sdp=True,
+                 **kwargs):
+
+        super().__init__()
+        self.n_vocab = n_vocab
+        self.spec_channels = spec_channels
+        self.inter_channels = inter_channels
+        self.hidden_channels = hidden_channels
+        self.filter_channels = filter_channels
+        self.n_heads = n_heads
+        self.n_layers = n_layers
+        self.kernel_size = kernel_size
+        self.p_dropout = p_dropout
+        self.resblock = resblock
+        self.resblock_kernel_sizes = resblock_kernel_sizes
+        self.resblock_dilation_sizes = resblock_dilation_sizes
+        self.upsample_rates = upsample_rates
+        self.upsample_initial_channel = upsample_initial_channel
+        self.upsample_kernel_sizes = upsample_kernel_sizes
+        self.segment_size = segment_size
+        self.n_speakers = n_speakers
+        self.gin_channels = gin_channels
+
+        self.use_sdp = use_sdp
+
+        self.enc_p = TextEncoder(n_vocab,
+                                 inter_channels,
+                                 hidden_channels,
+                                 filter_channels,
+                                 n_heads,
+                                 n_layers,
+                                 kernel_size,
+                                 p_dropout)
+        self.dec = Generator(inter_channels, resblock, resblock_kernel_sizes, resblock_dilation_sizes, upsample_rates,
+                             upsample_initial_channel, upsample_kernel_sizes, gin_channels=gin_channels)
+        self.enc_q = PosteriorEncoder(spec_channels, inter_channels, hidden_channels, 5, 1, 16,
+                                      gin_channels=gin_channels)
+        self.flow = ResidualCouplingBlock(inter_channels, hidden_channels, 5, 1, 4, gin_channels=gin_channels)
+
+        self.sdp = StochasticDurationPredictor(hidden_channels, 192, 3, 0.5, 4, gin_channels=gin_channels)
+        self.dp = DurationPredictor(hidden_channels, 256, 3, 0.5, gin_channels=gin_channels)
+
+        if n_speakers > 1:
+            self.emb_g = nn.Embedding(n_speakers, gin_channels)
+        else:
+            self.ref_enc = ReferenceEncoder()
+
+    def forward(self, x, x_lengths, y, y_lengths, sid, tone, language, bert):
+
+        x, m_p, logs_p, x_mask = self.enc_p(x, x_lengths, tone, language, bert)
+        if self.n_speakers > 0:
+            g = self.emb_g(sid).unsqueeze(-1)  # [b, h, 1]
+        else:
+            g = self.ref_enc(y.transpose(1,2)).unsqueeze(-1)
+
+        z, m_q, logs_q, y_mask = self.enc_q(y, y_lengths, g=g)
+        z_p = self.flow(z, y_mask, g=g)
+
+        with torch.no_grad():
+            # negative cross-entropy
+            s_p_sq_r = torch.exp(-2 * logs_p)  # [b, d, t]
+            neg_cent1 = torch.sum(-0.5 * math.log(2 * math.pi) - logs_p, [1], keepdim=True)  # [b, 1, t_s]
+            neg_cent2 = torch.matmul(-0.5 * (z_p ** 2).transpose(1, 2),
+                                     s_p_sq_r)  # [b, t_t, d] x [b, d, t_s] = [b, t_t, t_s]
+            neg_cent3 = torch.matmul(z_p.transpose(1, 2), (m_p * s_p_sq_r))  # [b, t_t, d] x [b, d, t_s] = [b, t_t, t_s]
+            neg_cent4 = torch.sum(-0.5 * (m_p ** 2) * s_p_sq_r, [1], keepdim=True)  # [b, 1, t_s]
+            neg_cent = neg_cent1 + neg_cent2 + neg_cent3 + neg_cent4
+
+            attn_mask = torch.unsqueeze(x_mask, 2) * torch.unsqueeze(y_mask, -1)
+            attn = monotonic_align.maximum_path(neg_cent, attn_mask.squeeze(1)).unsqueeze(1).detach()
+
+        w = attn.sum(2)
+
+        l_length_sdp = self.sdp(x, x_mask, w, g=g)
+        l_length_sdp = l_length_sdp / torch.sum(x_mask)
+
+        logw_ = torch.log(w + 1e-6) * x_mask
+        logw = self.dp(x, x_mask, g=g)
+        l_length_dp = torch.sum((logw - logw_) ** 2, [1, 2]) / torch.sum(x_mask)  # for averaging
+
+        l_length = l_length_dp + l_length_sdp
+
+        # expand prior
+        m_p = torch.matmul(attn.squeeze(1), m_p.transpose(1, 2)).transpose(1, 2)
+        logs_p = torch.matmul(attn.squeeze(1), logs_p.transpose(1, 2)).transpose(1, 2)
+
+        z_slice, ids_slice = commons.rand_slice_segments(z, y_lengths, self.segment_size)
+        o = self.dec(z_slice, g=g)
+        return o, l_length, attn, ids_slice, x_mask, y_mask, (z, z_p, m_p, logs_p, m_q, logs_q)
+
+    def infer(self, x, x_lengths, sid, tone, language, bert, noise_scale=.667, length_scale=1, noise_scale_w=0.8, max_len=None, sdp_ratio=0,y=None):
+        x, m_p, logs_p, x_mask = self.enc_p(x, x_lengths, tone, language, bert)
+        # g = self.gst(y)
+        if self.n_speakers > 0:
+            g = self.emb_g(sid).unsqueeze(-1)  # [b, h, 1]
+        else:
+            g = self.ref_enc(y.transpose(1,2)).unsqueeze(-1)
+
+        logw = self.sdp(x, x_mask, g=g, reverse=True, noise_scale=noise_scale_w) * (sdp_ratio) + self.dp(x, x_mask, g=g) * (1 - sdp_ratio)
+        w = torch.exp(logw) * x_mask * length_scale
+        w_ceil = torch.ceil(w)
+        y_lengths = torch.clamp_min(torch.sum(w_ceil, [1, 2]), 1).long()
+        y_mask = torch.unsqueeze(commons.sequence_mask(y_lengths, None), 1).to(x_mask.dtype)
+        attn_mask = torch.unsqueeze(x_mask, 2) * torch.unsqueeze(y_mask, -1)
+        attn = commons.generate_path(w_ceil, attn_mask)
+
+        m_p = torch.matmul(attn.squeeze(1), m_p.transpose(1, 2)).transpose(1, 2)  # [b, t', t], [b, t, d] -> [b, d, t']
+        logs_p = torch.matmul(attn.squeeze(1), logs_p.transpose(1, 2)).transpose(1,
+                                                                                 2)  # [b, t', t], [b, t, d] -> [b, d, t']
+
+        z_p = m_p + torch.randn_like(m_p) * torch.exp(logs_p) * noise_scale
+        z = self.flow(z_p, y_mask, g=g, reverse=True)
+        o = self.dec((z * y_mask)[:, :, :max_len], g=g)
+        return o, attn, y_mask, (z, z_p, m_p, logs_p)
+
+    def voice_conversion(self, y, y_lengths, sid_src, sid_tgt):
+        assert self.n_speakers > 0, "n_speakers have to be larger than 0."
+        g_src = self.emb_g(sid_src).unsqueeze(-1)
+        g_tgt = self.emb_g(sid_tgt).unsqueeze(-1)
+        z, m_q, logs_q, y_mask = self.enc_q(y, y_lengths, g=g_src)
+        z_p = self.flow(z, y_mask, g=g_src)
+        z_hat = self.flow(z_p, y_mask, g=g_tgt, reverse=True)
+        o_hat = self.dec(z_hat * y_mask, g=g_tgt)
+        return o_hat, y_mask, (z, z_p, z_hat)

modules.py

@@ -0,0 +1,390 @@
+import copy
+import math
+import numpy as np
+import scipy
+import torch
+from torch import nn
+from torch.nn import functional as F
+
+from torch.nn import Conv1d, ConvTranspose1d, AvgPool1d, Conv2d
+from torch.nn.utils import weight_norm, remove_weight_norm
+
+import commons
+from commons import init_weights, get_padding
+from transforms import piecewise_rational_quadratic_transform
+
+
+LRELU_SLOPE = 0.1
+
+
+class LayerNorm(nn.Module):
+  def __init__(self, channels, eps=1e-5):
+    super().__init__()
+    self.channels = channels
+    self.eps = eps
+
+    self.gamma = nn.Parameter(torch.ones(channels))
+    self.beta = nn.Parameter(torch.zeros(channels))
+
+  def forward(self, x):
+    x = x.transpose(1, -1)
+    x = F.layer_norm(x, (self.channels,), self.gamma, self.beta, self.eps)
+    return x.transpose(1, -1)
+
+ 
+class ConvReluNorm(nn.Module):
+  def __init__(self, in_channels, hidden_channels, out_channels, kernel_size, n_layers, p_dropout):
+    super().__init__()
+    self.in_channels = in_channels
+    self.hidden_channels = hidden_channels
+    self.out_channels = out_channels
+    self.kernel_size = kernel_size
+    self.n_layers = n_layers
+    self.p_dropout = p_dropout
+    assert n_layers > 1, "Number of layers should be larger than 0."
+
+    self.conv_layers = nn.ModuleList()
+    self.norm_layers = nn.ModuleList()
+    self.conv_layers.append(nn.Conv1d(in_channels, hidden_channels, kernel_size, padding=kernel_size//2))
+    self.norm_layers.append(LayerNorm(hidden_channels))
+    self.relu_drop = nn.Sequential(
+        nn.ReLU(),
+        nn.Dropout(p_dropout))
+    for _ in range(n_layers-1):
+      self.conv_layers.append(nn.Conv1d(hidden_channels, hidden_channels, kernel_size, padding=kernel_size//2))
+      self.norm_layers.append(LayerNorm(hidden_channels))
+    self.proj = nn.Conv1d(hidden_channels, out_channels, 1)
+    self.proj.weight.data.zero_()
+    self.proj.bias.data.zero_()
+
+  def forward(self, x, x_mask):
+    x_org = x
+    for i in range(self.n_layers):
+      x = self.conv_layers[i](x * x_mask)
+      x = self.norm_layers[i](x)
+      x = self.relu_drop(x)
+    x = x_org + self.proj(x)
+    return x * x_mask
+
+
+class DDSConv(nn.Module):
+  """
+  Dialted and Depth-Separable Convolution
+  """
+  def __init__(self, channels, kernel_size, n_layers, p_dropout=0.):
+    super().__init__()
+    self.channels = channels
+    self.kernel_size = kernel_size
+    self.n_layers = n_layers
+    self.p_dropout = p_dropout
+
+    self.drop = nn.Dropout(p_dropout)
+    self.convs_sep = nn.ModuleList()
+    self.convs_1x1 = nn.ModuleList()
+    self.norms_1 = nn.ModuleList()
+    self.norms_2 = nn.ModuleList()
+    for i in range(n_layers):
+      dilation = kernel_size ** i
+      padding = (kernel_size * dilation - dilation) // 2
+      self.convs_sep.append(nn.Conv1d(channels, channels, kernel_size, 
+          groups=channels, dilation=dilation, padding=padding
+      ))
+      self.convs_1x1.append(nn.Conv1d(channels, channels, 1))
+      self.norms_1.append(LayerNorm(channels))
+      self.norms_2.append(LayerNorm(channels))
+
+  def forward(self, x, x_mask, g=None):
+    if g is not None:
+      x = x + g
+    for i in range(self.n_layers):
+      y = self.convs_sep[i](x * x_mask)
+      y = self.norms_1[i](y)
+      y = F.gelu(y)
+      y = self.convs_1x1[i](y)
+      y = self.norms_2[i](y)
+      y = F.gelu(y)
+      y = self.drop(y)
+      x = x + y
+    return x * x_mask
+
+
+class WN(torch.nn.Module):
+  def __init__(self, hidden_channels, kernel_size, dilation_rate, n_layers, gin_channels=0, p_dropout=0):
+    super(WN, self).__init__()
+    assert(kernel_size % 2 == 1)
+    self.hidden_channels =hidden_channels
+    self.kernel_size = kernel_size,
+    self.dilation_rate = dilation_rate
+    self.n_layers = n_layers
+    self.gin_channels = gin_channels
+    self.p_dropout = p_dropout
+
+    self.in_layers = torch.nn.ModuleList()
+    self.res_skip_layers = torch.nn.ModuleList()
+    self.drop = nn.Dropout(p_dropout)
+
+    if gin_channels != 0:
+      cond_layer = torch.nn.Conv1d(gin_channels, 2*hidden_channels*n_layers, 1)
+      self.cond_layer = torch.nn.utils.weight_norm(cond_layer, name='weight')
+
+    for i in range(n_layers):
+      dilation = dilation_rate ** i
+      padding = int((kernel_size * dilation - dilation) / 2)
+      in_layer = torch.nn.Conv1d(hidden_channels, 2*hidden_channels, kernel_size,
+                                 dilation=dilation, padding=padding)
+      in_layer = torch.nn.utils.weight_norm(in_layer, name='weight')
+      self.in_layers.append(in_layer)
+
+      # last one is not necessary
+      if i < n_layers - 1:
+        res_skip_channels = 2 * hidden_channels
+      else:
+        res_skip_channels = hidden_channels
+
+      res_skip_layer = torch.nn.Conv1d(hidden_channels, res_skip_channels, 1)
+      res_skip_layer = torch.nn.utils.weight_norm(res_skip_layer, name='weight')
+      self.res_skip_layers.append(res_skip_layer)
+
+  def forward(self, x, x_mask, g=None, **kwargs):
+    output = torch.zeros_like(x)
+    n_channels_tensor = torch.IntTensor([self.hidden_channels])
+
+    if g is not None:
+      g = self.cond_layer(g)
+
+    for i in range(self.n_layers):
+      x_in = self.in_layers[i](x)
+      if g is not None:
+        cond_offset = i * 2 * self.hidden_channels
+        g_l = g[:,cond_offset:cond_offset+2*self.hidden_channels,:]
+      else:
+        g_l = torch.zeros_like(x_in)
+
+      acts = commons.fused_add_tanh_sigmoid_multiply(
+          x_in,
+          g_l,
+          n_channels_tensor)
+      acts = self.drop(acts)
+
+      res_skip_acts = self.res_skip_layers[i](acts)
+      if i < self.n_layers - 1:
+        res_acts = res_skip_acts[:,:self.hidden_channels,:]
+        x = (x + res_acts) * x_mask
+        output = output + res_skip_acts[:,self.hidden_channels:,:]
+      else:
+        output = output + res_skip_acts
+    return output * x_mask
+
+  def remove_weight_norm(self):
+    if self.gin_channels != 0:
+      torch.nn.utils.remove_weight_norm(self.cond_layer)
+    for l in self.in_layers:
+      torch.nn.utils.remove_weight_norm(l)
+    for l in self.res_skip_layers:
+     torch.nn.utils.remove_weight_norm(l)
+
+
+class ResBlock1(torch.nn.Module):
+    def __init__(self, channels, kernel_size=3, dilation=(1, 3, 5)):
+        super(ResBlock1, self).__init__()
+        self.convs1 = nn.ModuleList([
+            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=dilation[0],
+                               padding=get_padding(kernel_size, dilation[0]))),
+            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=dilation[1],
+                               padding=get_padding(kernel_size, dilation[1]))),
+            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=dilation[2],
+                               padding=get_padding(kernel_size, dilation[2])))
+        ])
+        self.convs1.apply(init_weights)
+
+        self.convs2 = nn.ModuleList([
+            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=1,
+                               padding=get_padding(kernel_size, 1))),
+            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=1,
+                               padding=get_padding(kernel_size, 1))),
+            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=1,
+                               padding=get_padding(kernel_size, 1)))
+        ])
+        self.convs2.apply(init_weights)
+
+    def forward(self, x, x_mask=None):
+        for c1, c2 in zip(self.convs1, self.convs2):
+            xt = F.leaky_relu(x, LRELU_SLOPE)
+            if x_mask is not None:
+                xt = xt * x_mask
+            xt = c1(xt)
+            xt = F.leaky_relu(xt, LRELU_SLOPE)
+            if x_mask is not None:
+                xt = xt * x_mask
+            xt = c2(xt)
+            x = xt + x
+        if x_mask is not None:
+            x = x * x_mask
+        return x
+
+    def remove_weight_norm(self):
+        for l in self.convs1:
+            remove_weight_norm(l)
+        for l in self.convs2:
+            remove_weight_norm(l)
+
+
+class ResBlock2(torch.nn.Module):
+    def __init__(self, channels, kernel_size=3, dilation=(1, 3)):
+        super(ResBlock2, self).__init__()
+        self.convs = nn.ModuleList([
+            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=dilation[0],
+                               padding=get_padding(kernel_size, dilation[0]))),
+            weight_norm(Conv1d(channels, channels, kernel_size, 1, dilation=dilation[1],
+                               padding=get_padding(kernel_size, dilation[1])))
+        ])
+        self.convs.apply(init_weights)
+
+    def forward(self, x, x_mask=None):
+        for c in self.convs:
+            xt = F.leaky_relu(x, LRELU_SLOPE)
+            if x_mask is not None:
+                xt = xt * x_mask
+            xt = c(xt)
+            x = xt + x
+        if x_mask is not None:
+            x = x * x_mask
+        return x
+
+    def remove_weight_norm(self):
+        for l in self.convs:
+            remove_weight_norm(l)
+
+
+class Log(nn.Module):
+  def forward(self, x, x_mask, reverse=False, **kwargs):
+    if not reverse:
+      y = torch.log(torch.clamp_min(x, 1e-5)) * x_mask
+      logdet = torch.sum(-y, [1, 2])
+      return y, logdet
+    else:
+      x = torch.exp(x) * x_mask
+      return x
+    
+
+class Flip(nn.Module):
+  def forward(self, x, *args, reverse=False, **kwargs):
+    x = torch.flip(x, [1])
+    if not reverse:
+      logdet = torch.zeros(x.size(0)).to(dtype=x.dtype, device=x.device)
+      return x, logdet
+    else:
+      return x
+
+
+class ElementwiseAffine(nn.Module):
+  def __init__(self, channels):
+    super().__init__()
+    self.channels = channels
+    self.m = nn.Parameter(torch.zeros(channels,1))
+    self.logs = nn.Parameter(torch.zeros(channels,1))
+
+  def forward(self, x, x_mask, reverse=False, **kwargs):
+    if not reverse:
+      y = self.m + torch.exp(self.logs) * x
+      y = y * x_mask
+      logdet = torch.sum(self.logs * x_mask, [1,2])
+      return y, logdet
+    else:
+      x = (x - self.m) * torch.exp(-self.logs) * x_mask
+      return x
+
+
+class ResidualCouplingLayer(nn.Module):
+  def __init__(self,
+      channels,
+      hidden_channels,
+      kernel_size,
+      dilation_rate,
+      n_layers,
+      p_dropout=0,
+      gin_channels=0,
+      mean_only=False):
+    assert channels % 2 == 0, "channels should be divisible by 2"
+    super().__init__()
+    self.channels = channels
+    self.hidden_channels = hidden_channels
+    self.kernel_size = kernel_size
+    self.dilation_rate = dilation_rate
+    self.n_layers = n_layers
+    self.half_channels = channels // 2
+    self.mean_only = mean_only
+
+    self.pre = nn.Conv1d(self.half_channels, hidden_channels, 1)
+    self.enc = WN(hidden_channels, kernel_size, dilation_rate, n_layers, p_dropout=p_dropout, gin_channels=gin_channels)
+    self.post = nn.Conv1d(hidden_channels, self.half_channels * (2 - mean_only), 1)
+    self.post.weight.data.zero_()
+    self.post.bias.data.zero_()
+
+  def forward(self, x, x_mask, g=None, reverse=False):
+    x0, x1 = torch.split(x, [self.half_channels]*2, 1)
+    h = self.pre(x0) * x_mask
+    h = self.enc(h, x_mask, g=g)
+    stats = self.post(h) * x_mask
+    if not self.mean_only:
+      m, logs = torch.split(stats, [self.half_channels]*2, 1)
+    else:
+      m = stats
+      logs = torch.zeros_like(m)
+
+    if not reverse:
+      x1 = m + x1 * torch.exp(logs) * x_mask
+      x = torch.cat([x0, x1], 1)
+      logdet = torch.sum(logs, [1,2])
+      return x, logdet
+    else:
+      x1 = (x1 - m) * torch.exp(-logs) * x_mask
+      x = torch.cat([x0, x1], 1)
+      return x
+
+
+class ConvFlow(nn.Module):
+  def __init__(self, in_channels, filter_channels, kernel_size, n_layers, num_bins=10, tail_bound=5.0):
+    super().__init__()
+    self.in_channels = in_channels
+    self.filter_channels = filter_channels
+    self.kernel_size = kernel_size
+    self.n_layers = n_layers
+    self.num_bins = num_bins
+    self.tail_bound = tail_bound
+    self.half_channels = in_channels // 2
+
+    self.pre = nn.Conv1d(self.half_channels, filter_channels, 1)
+    self.convs = DDSConv(filter_channels, kernel_size, n_layers, p_dropout=0.)
+    self.proj = nn.Conv1d(filter_channels, self.half_channels * (num_bins * 3 - 1), 1)
+    self.proj.weight.data.zero_()
+    self.proj.bias.data.zero_()
+
+  def forward(self, x, x_mask, g=None, reverse=False):
+    x0, x1 = torch.split(x, [self.half_channels]*2, 1)
+    h = self.pre(x0)
+    h = self.convs(h, x_mask, g=g)
+    h = self.proj(h) * x_mask
+
+    b, c, t = x0.shape
+    h = h.reshape(b, c, -1, t).permute(0, 1, 3, 2) # [b, cx?, t] -> [b, c, t, ?]
+
+    unnormalized_widths = h[..., :self.num_bins] / math.sqrt(self.filter_channels)
+    unnormalized_heights = h[..., self.num_bins:2*self.num_bins] / math.sqrt(self.filter_channels)
+    unnormalized_derivatives = h[..., 2 * self.num_bins:]
+
+    x1, logabsdet = piecewise_rational_quadratic_transform(x1,
+        unnormalized_widths,
+        unnormalized_heights,
+        unnormalized_derivatives,
+        inverse=reverse,
+        tails='linear',
+        tail_bound=self.tail_bound
+    )
+
+    x = torch.cat([x0, x1], 1) * x_mask
+    logdet = torch.sum(logabsdet * x_mask, [1,2])
+    if not reverse:
+        return x, logdet
+    else:
+        return x

monotonic_align/__init__.py

@@ -0,0 +1,19 @@
+import numpy as np
+import torch
+from .monotonic_align.core import maximum_path_c
+
+
+def maximum_path(neg_cent, mask):
+  """ Cython optimized version.
+  neg_cent: [b, t_t, t_s]
+  mask: [b, t_t, t_s]
+  """
+  device = neg_cent.device
+  dtype = neg_cent.dtype
+  neg_cent = neg_cent.data.cpu().numpy().astype(np.float32)
+  path = np.zeros(neg_cent.shape, dtype=np.int32)
+
+  t_t_max = mask.sum(1)[:, 0].data.cpu().numpy().astype(np.int32)
+  t_s_max = mask.sum(2)[:, 0].data.cpu().numpy().astype(np.int32)
+  maximum_path_c(path, neg_cent, t_t_max, t_s_max)
+  return torch.from_numpy(path).to(device=device, dtype=dtype)

monotonic_align/core.pyx

@@ -0,0 +1,42 @@
+cimport cython
+from cython.parallel import prange
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+cdef void maximum_path_each(int[:,::1] path, float[:,::1] value, int t_y, int t_x, float max_neg_val=-1e9) nogil:
+  cdef int x
+  cdef int y
+  cdef float v_prev
+  cdef float v_cur
+  cdef float tmp
+  cdef int index = t_x - 1
+
+  for y in range(t_y):
+    for x in range(max(0, t_x + y - t_y), min(t_x, y + 1)):
+      if x == y:
+        v_cur = max_neg_val
+      else:
+        v_cur = value[y-1, x]
+      if x == 0:
+        if y == 0:
+          v_prev = 0.
+        else:
+          v_prev = max_neg_val
+      else:
+        v_prev = value[y-1, x-1]
+      value[y, x] += max(v_prev, v_cur)
+
+  for y in range(t_y - 1, -1, -1):
+    path[y, index] = 1
+    if index != 0 and (index == y or value[y-1, index] < value[y-1, index-1]):
+      index = index - 1
+
+
+@cython.boundscheck(False)
+@cython.wraparound(False)
+cpdef void maximum_path_c(int[:,:,::1] paths, float[:,:,::1] values, int[::1] t_ys, int[::1] t_xs) nogil:
+  cdef int b = paths.shape[0]
+  cdef int i
+  for i in prange(b, nogil=True):
+    maximum_path_each(paths[i], values[i], t_ys[i], t_xs[i])

monotonic_align/setup.py

@@ -0,0 +1,9 @@
+from distutils.core import setup
+from Cython.Build import cythonize
+import numpy
+
+setup(
+  name = 'monotonic_align',
+  ext_modules = cythonize("core.pyx"),
+  include_dirs=[numpy.get_include()]
+)

preprocess.py

@@ -0,0 +1,25 @@
+import argparse
+import text
+from utils import load_filepaths_and_text
+
+if __name__ == '__main__':
+  parser = argparse.ArgumentParser()
+  parser.add_argument("--out_extension", default="cleaned")
+  parser.add_argument("--text_index", default=1, type=int)
+  parser.add_argument("--filelists", nargs="+", default=["filelists/ljs_audio_text_val_filelist.txt", "filelists/ljs_audio_text_test_filelist.txt"])
+  parser.add_argument("--text_cleaners", nargs="+", default=["english_cleaners2"])
+
+  args = parser.parse_args()
+    
+
+  for filelist in args.filelists:
+    print("START:", filelist)
+    filepaths_and_text = load_filepaths_and_text(filelist)
+    for i in range(len(filepaths_and_text)):
+      original_text = filepaths_and_text[i][args.text_index]
+      cleaned_text = text._clean_text(original_text, args.text_cleaners)
+      filepaths_and_text[i][args.text_index] = cleaned_text
+
+    new_filelist = filelist + "." + args.out_extension
+    with open(new_filelist, "w", encoding="utf-8") as f:
+      f.writelines(["|".join(x) + "\n" for x in filepaths_and_text])

preprocess_text.py

@@ -0,0 +1,62 @@
+import json
+from random import shuffle
+
+import tqdm
+from text.cleaner import clean_text
+from collections import defaultdict
+stage = [1,2,3]
+
+transcription_path = 'filelists/esd.list'
+train_path = 'filelists/train.list'
+val_path = 'filelists/val.list'
+config_path = "configs/config.json"
+val_per_spk = 2
+max_val_total = 8
+
+if 1 in stage:
+    with open( transcription_path+'.cleaned', 'w', encoding='utf-8') as f:
+        for line in tqdm.tqdm(open(transcription_path, encoding='utf-8').readlines()):
+            utt, spk, language, text = line.strip().split('|')
+            norm_text, phones, tones, word2ph = clean_text(text, language)
+            f.write('{}|{}|{}|{}|{}|{}|{}\n'.format(utt, spk, language, norm_text, ' '.join(phones),
+                                                 " ".join([str(i) for i in tones]),
+                                                 " ".join([str(i) for i in word2ph])))
+
+if 2 in stage:
+    spk_utt_map = defaultdict(list)
+    spk_id_map = {}
+    current_sid = 0
+
+    with open( transcription_path+'.cleaned', encoding='utf-8') as f:
+        for line in f.readlines():
+            utt, spk, language, text, phones, tones, word2ph = line.strip().split('|')
+            spk_utt_map[spk].append(line)
+            if spk not in spk_id_map.keys():
+                spk_id_map[spk] = current_sid
+                current_sid += 1
+
+    train_list = []
+    val_list = []
+
+    for spk, utts in spk_utt_map.items():
+        shuffle(utts)
+        val_list+=utts[:val_per_spk]
+        train_list+=utts[val_per_spk:]
+    if len(val_list) > max_val_total:
+        train_list+=val_list[max_val_total:]
+        val_list = val_list[:max_val_total]
+
+    with open( train_path,"w", encoding='utf-8') as f:
+        for line in train_list:
+            f.write(line)
+
+    with open(val_path, "w", encoding='utf-8') as f:
+        for line in val_list:
+            f.write(line)
+
+if 3 in stage:
+    assert 2 in stage
+    config = json.load(open(config_path))
+    config["data"]['spk2id'] = spk_id_map
+    with open(config_path, 'w', encoding='utf-8') as f:
+        json.dump(config, f, indent=2)

requirements.txt

@@ -0,0 +1,11 @@
+Cython==0.29.21
+librosa==0.8.0
+matplotlib==3.3.1
+numpy==1.18.5
+phonemizer==2.2.1
+scipy==1.5.2
+tensorboard==2.3.0
+torch==1.6.0
+torchvision==0.7.0
+Unidecode==1.1.1
+amfm_decompy

resample.py

@@ -0,0 +1,42 @@
+import os
+import argparse
+import librosa
+import numpy as np
+from multiprocessing import Pool, cpu_count
+
+import soundfile
+from scipy.io import wavfile
+from tqdm import tqdm
+
+
+def process(item):
+    spkdir, wav_name, args = item
+    # speaker 's5', 'p280', 'p315' are excluded,
+    speaker = spkdir.replace("\\", "/").split("/")[-1]
+    wav_path = os.path.join(args.in_dir, speaker, wav_name)
+    if os.path.exists(wav_path) and '.wav' in wav_path:
+        os.makedirs(os.path.join(args.out_dir2, speaker), exist_ok=True)
+        wav, sr = librosa.load(wav_path, sr=args.sr2)
+        soundfile.write(
+            os.path.join(args.out_dir2, speaker, wav_name),
+            wav,
+            sr
+        )
+
+
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--sr2", type=int, default=22050, help="sampling rate")
+    parser.add_argument("--in_dir", type=str, default="./raw", help="path to source dir")
+    parser.add_argument("--out_dir2", type=str, default="./dataset", help="path to target dir")
+    args = parser.parse_args()
+    processs = 8
+    pool = Pool(processes=processs)
+
+    for speaker in os.listdir(args.in_dir):
+        spk_dir = os.path.join(args.in_dir, speaker)
+        if os.path.isdir(spk_dir):
+            print(spk_dir)
+            for _ in tqdm(pool.imap_unordered(process, [(spk_dir, i, args) for i in os.listdir(spk_dir) if i.endswith("wav")])):
+                pass

text/__init__.py

@@ -0,0 +1,26 @@
+from text.symbols import *
+
+
+_symbol_to_id = {s: i for i, s in enumerate(symbols)}
+
+def cleaned_text_to_sequence(cleaned_text, tones, language):
+  '''Converts a string of text to a sequence of IDs corresponding to the symbols in the text.
+    Args:
+      text: string to convert to a sequence
+    Returns:
+      List of integers corresponding to the symbols in the text
+  '''
+  phones = [_symbol_to_id[symbol] for symbol in cleaned_text]
+  tone_start = language_tone_start_map[language]
+  tones = [i + tone_start for i in tones]
+  lang_id = language_id_map[language]
+  lang_ids = [lang_id for i in phones]
+  return phones, tones, lang_ids
+
+def get_bert(norm_text, word2ph, language):
+  from chinese_bert import get_bert_feature as zh_bert
+  lang_bert_func_map = {
+    'ZH': zh_bert
+  }
+  bert = lang_bert_func_map[language](norm_text, word2ph)
+  return bert

text/chinese.py

@@ -0,0 +1,193 @@
+import os
+import re
+
+import cn2an
+from pypinyin import lazy_pinyin, Style
+
+from text import symbols
+from text.symbols import punctuation
+from text.tone_sandhi import ToneSandhi
+
+current_file_path = os.path.dirname(__file__)
+pinyin_to_symbol_map = {line.split("\t")[0]: line.strip().split("\t")[1] for line in
+                        open(os.path.join(current_file_path, 'opencpop-strict.txt')).readlines()}
+
+import jieba.posseg as psg
+
+
+rep_map = {
+    '：': ',',
+    '；': ',',
+    '，': ',',
+    '。': '.',
+    '！': '!',
+    '？': '?',
+    '\n': '.',
+    "·": ",",
+    '、': ",",
+    '...': '…',
+    '$': '.',
+    '“': "'",
+    '”': "'",
+    '‘': "'",
+    '’': "'",
+    '（': "'",
+    '）': "'",
+    '(': "'",
+    ')': "'",
+    '《': "'",
+    '》': "'",
+    '【': "'",
+    '】': "'",
+    '[': "'",
+    ']': "'",
+    '—': "-",
+    '～': "-",
+    '~': "-",
+    '「': "'",
+    '」': "'",
+
+}
+
+tone_modifier = ToneSandhi()
+
+def replace_punctuation(text):
+    text = text.replace("嗯", "恩").replace("呣","母")
+    pattern = re.compile('|'.join(re.escape(p) for p in rep_map.keys()))
+
+    replaced_text = pattern.sub(lambda x: rep_map[x.group()], text)
+
+    replaced_text = re.sub(r'[^\u4e00-\u9fa5'+"".join(punctuation)+r']+', '', replaced_text)
+
+    return replaced_text
+
+def g2p(text):
+    pattern = r'(?<=[{0}])\s*'.format(''.join(punctuation))
+    sentences = [i for i in re.split(pattern, text) if i.strip()!='']
+    phones, tones, word2ph = _g2p(sentences)
+    assert sum(word2ph) == len(phones)
+    assert len(word2ph) == len(text)
+    phones = ['_'] + phones + ["_"]
+    tones = [0] + tones + [0]
+    word2ph = [1] + word2ph + [1]
+    return phones, tones, word2ph
+
+
+def _get_initials_finals(word):
+    initials = []
+    finals = []
+    orig_initials = lazy_pinyin(
+        word, neutral_tone_with_five=True, style=Style.INITIALS)
+    orig_finals = lazy_pinyin(
+        word, neutral_tone_with_five=True, style=Style.FINALS_TONE3)
+    for c, v in zip(orig_initials, orig_finals):
+        initials.append(c)
+        finals.append(v)
+    return initials, finals
+
+
+def _g2p(segments):
+    phones_list = []
+    tones_list = []
+    word2ph = []
+    for seg in segments:
+        pinyins = []
+        # Replace all English words in the sentence
+        seg = re.sub('[a-zA-Z]+', '', seg)
+        seg_cut = psg.lcut(seg)
+        initials = []
+        finals = []
+        seg_cut = tone_modifier.pre_merge_for_modify(seg_cut)
+        for word, pos in seg_cut:
+            if pos == 'eng':
+                continue
+            sub_initials, sub_finals = _get_initials_finals(word)
+            sub_finals = tone_modifier.modified_tone(word, pos,
+                                                          sub_finals)
+            initials.append(sub_initials)
+            finals.append(sub_finals)
+
+            # assert len(sub_initials) == len(sub_finals) == len(word)
+        initials = sum(initials, [])
+        finals = sum(finals, [])
+        #
+        for c, v in zip(initials, finals):
+            raw_pinyin = c+v
+            # NOTE: post process for pypinyin outputs
+            # we discriminate i, ii and iii
+            if c == v:
+                assert c in punctuation
+                phone = [c]
+                tone = '0'
+                word2ph.append(1)
+            else:
+                v_without_tone = v[:-1]
+                tone = v[-1]
+
+                pinyin = c+v_without_tone
+                assert tone in '12345'
+
+                if c:
+                    # 多音节
+                    v_rep_map = {
+                        "uei": 'ui',
+                        'iou': 'iu',
+                        'uen': 'un',
+                    }
+                    if v_without_tone in v_rep_map.keys():
+                        pinyin = c+v_rep_map[v_without_tone]
+                else:
+                    # 单音节
+                    pinyin_rep_map = {
+                        'ing': 'ying',
+                        'i': 'yi',
+                        'in': 'yin',
+                        'u': 'wu',
+                    }
+                    if pinyin in pinyin_rep_map.keys():
+                        pinyin = pinyin_rep_map[pinyin]
+                    else:
+                        single_rep_map = {
+                            'v': 'yu',
+                            'e': 'e',
+                            'i': 'y',
+                            'u': 'w',
+                        }
+                        if pinyin[0] in single_rep_map.keys():
+                            pinyin = single_rep_map[pinyin[0]]+pinyin[1:]
+
+                assert pinyin in pinyin_to_symbol_map.keys(), (pinyin, seg, raw_pinyin)
+                phone = pinyin_to_symbol_map[pinyin].split(' ')
+                word2ph.append(len(phone))
+
+            phones_list += phone
+            tones_list += [int(tone)] * len(phone)
+    return phones_list, tones_list, word2ph
+
+
+
+def text_normalize(text):
+    numbers = re.findall(r'\d+(?:\.?\d+)?', text)
+    for number in numbers:
+        text = text.replace(number, cn2an.an2cn(number), 1)
+    text = replace_punctuation(text)
+    return text
+
+def get_bert_feature(text, word2ph):
+    from  text import chinese_bert
+    return chinese_bert.get_bert_feature(text, word2ph)
+
+if __name__ == '__main__':
+    from text.chinese_bert import get_bert_feature
+    text = "啊！但是《原神》是由,米哈\游自主，  [研发]的一款全.新开放世界.冒险游戏"
+    text = text_normalize(text)
+    print(text)
+    phones, tones, word2ph = g2p(text)
+    bert = get_bert_feature(text, word2ph)
+
+    print(phones, tones, word2ph, bert.shape)
+
+
+# # 示例用法
+# text = "这是一个示例文本：,你好！这是一个测试...."
+# print(g2p_paddle(text))  # 输出: 这是一个示例文本你好这是一个测试

text/chinese_bert.py

@@ -0,0 +1,55 @@
+import torch
+from transformers import AutoTokenizer, AutoModelForMaskedLM
+import os
+
+os.environ['ALL_PROXY']='socks5://127.0.0.1:7890'
+os.environ['HTTPS_PROXY']='http://127.0.0.1:7890'
+os.environ['HTTP_PROXY']='http://127.0.0.1:7890'
+
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+tokenizer = AutoTokenizer.from_pretrained("hfl/chinese-roberta-wwm-ext-large")
+model = AutoModelForMaskedLM.from_pretrained("hfl/chinese-roberta-wwm-ext-large").to(device)
+
+def get_bert_feature(text, word2ph):
+    with torch.no_grad():
+        inputs = tokenizer(text, return_tensors='pt')
+        for i in inputs:
+            inputs[i] = inputs[i].to(device)
+        res = model(**inputs, output_hidden_states=True)
+        res = torch.cat(res['hidden_states'][-3:-2], -1)[0].cpu()
+
+    assert len(word2ph) == len(text)+2
+    word2phone = word2ph
+    phone_level_feature = []
+    for i in range(len(word2phone)):
+        repeat_feature = res[i].repeat(word2phone[i], 1)
+        phone_level_feature.append(repeat_feature)
+
+    phone_level_feature = torch.cat(phone_level_feature, dim=0)
+
+
+    return phone_level_feature.T
+
+if __name__ == '__main__':
+    # feature = get_bert_feature('你好,我是说的道理。')
+    import torch
+
+    word_level_feature = torch.rand(38, 1024)  # 12个词,每个词1024维特征
+    word2phone = [1, 2, 1, 2, 2, 1, 2, 2, 1, 2, 2, 1, 2, 2, 2, 2, 2, 1, 1, 2, 2, 1, 2, 2, 2, 2, 1, 2, 2, 2, 2, 2, 1, 2, 2, 2, 2, 1]
+
+    # 计算总帧数
+    total_frames = sum(word2phone)
+    print(word_level_feature.shape)
+    print(word2phone)
+    phone_level_feature = []
+    for i in range(len(word2phone)):
+        print(word_level_feature[i].shape)
+
+        # 对每个词重复word2phone[i]次
+        repeat_feature = word_level_feature[i].repeat(word2phone[i], 1)
+        phone_level_feature.append(repeat_feature)
+
+    phone_level_feature = torch.cat(phone_level_feature, dim=0)
+    print(phone_level_feature.shape)  # torch.Size([36, 1024])
+

text/cleaner.py

@@ -0,0 +1,32 @@
+from text import chinese, japanese, english, cleaned_text_to_sequence
+
+
+language_module_map = {
+    'ZH': chinese,
+    "JA": japanese,
+    "EN": english
+}
+
+
+def clean_text(text, language):
+    language_module = language_module_map[language]
+    norm_text = language_module.text_normalize(text)
+    phones, tones, word2ph = language_module.g2p(norm_text)
+    return norm_text, phones, tones, word2ph
+
+def clean_text_bert(text, language):
+    language_module = language_module_map[language]
+    norm_text = language_module.text_normalize(text)
+    phones, tones, word2ph = language_module.g2p(norm_text)
+    bert = language_module.get_bert_feature(norm_text, word2ph)
+    return phones, tones, bert
+
+def text_to_sequence(text, language):
+    _, _, phones, tones = clean_text(text, language)
+    return cleaned_text_to_sequence(phones, tones, language)
+
+if __name__ == '__main__':
+    print(text_to_sequence("你好，啊啊啊额、还是到付红四方。", 'ZH'))
+    print(text_to_sequence("hello", 'EN'))
+
+

text/english.py

@@ -0,0 +1,138 @@
+import pickle
+import os
+import re
+from g2p_en import G2p
+from string import punctuation
+
+from text import symbols
+
+current_file_path = os.path.dirname(__file__)
+CMU_DICT_PATH = os.path.join(current_file_path, 'cmudict.rep')
+CACHE_PATH = os.path.join(current_file_path, 'cmudict_cache.pickle')
+_g2p = G2p()
+
+arpa = {'AH0', 'S', 'AH1', 'EY2', 'AE2', 'EH0', 'OW2', 'UH0', 'NG', 'B', 'G', 'AY0', 'M', 'AA0', 'F', 'AO0', 'ER2', 'UH1', 'IY1', 'AH2', 'DH', 'IY0', 'EY1', 'IH0', 'K', 'N', 'W', 'IY2', 'T', 'AA1', 'ER1', 'EH2', 'OY0', 'UH2', 'UW1', 'Z', 'AW2', 'AW1', 'V', 'UW2', 'AA2', 'ER', 'AW0', 'UW0', 'R', 'OW1', 'EH1', 'ZH', 'AE0', 'IH2', 'IH', 'Y', 'JH', 'P', 'AY1', 'EY0', 'OY2', 'TH', 'HH', 'D', 'ER0', 'CH', 'AO1', 'AE1', 'AO2', 'OY1', 'AY2', 'IH1', 'OW0', 'L', 'SH'}
+
+
+def post_replace_ph(ph):
+    rep_map = {
+        '：': ',',
+        '；': ',',
+        '，': ',',
+        '。': '.',
+        '！': '!',
+        '？': '?',
+        '\n': '.',
+        "·": ",",
+        '、': ",",
+        '...': '…',
+        'v': "V"
+    }
+    if ph in rep_map.keys():
+        ph = rep_map[ph]
+    if ph in symbols:
+        return ph
+    if ph not in symbols:
+        ph = 'UNK'
+    return ph
+
+def read_dict():
+    g2p_dict = {}
+    start_line = 49
+    with open(CMU_DICT_PATH) as f:
+        line = f.readline()
+        line_index = 1
+        while line:
+            if line_index >= start_line:
+                line = line.strip()
+                word_split = line.split('  ')
+                word = word_split[0]
+
+                syllable_split = word_split[1].split(' - ')
+                g2p_dict[word] = []
+                for syllable in syllable_split:
+                    phone_split = syllable.split(' ')
+                    g2p_dict[word].append(phone_split)
+
+            line_index = line_index + 1
+            line = f.readline()
+
+    return g2p_dict
+
+
+def cache_dict(g2p_dict, file_path):
+    with open(file_path, 'wb') as pickle_file:
+        pickle.dump(g2p_dict, pickle_file)
+
+
+def get_dict():
+    if os.path.exists(CACHE_PATH):
+        with open(CACHE_PATH, 'rb') as pickle_file:
+            g2p_dict = pickle.load(pickle_file)
+    else:
+        g2p_dict = read_dict()
+        cache_dict(g2p_dict, CACHE_PATH)
+
+    return g2p_dict
+
+eng_dict = get_dict()
+
+def refine_ph(phn):
+    tone = 0
+    if re.search(r'\d$', phn):
+        tone = int(phn[-1]) + 1
+        phn = phn[:-1]
+    return phn.lower(), tone
+
+def refine_syllables(syllables):
+    tones = []
+    phonemes = []
+    for phn_list in syllables:
+        for i in range(len(phn_list)):
+            phn = phn_list[i]
+            phn, tone = refine_ph(phn)
+            phonemes.append(phn)
+            tones.append(tone)
+    return phonemes, tones
+
+
+def text_normalize(text):
+    # todo: eng text normalize
+    return text
+
+def g2p(text):
+
+    phones = []
+    tones = []
+    words = re.split(r"([,;.\-\?\!\s+])", text)
+    for w in words:
+        if w.upper() in eng_dict:
+            phns, tns = refine_syllables(eng_dict[w.upper()])
+            phones += phns
+            tones += tns
+        else:
+            phone_list = list(filter(lambda p: p != " ", _g2p(w)))
+            for ph in phone_list:
+                if ph in arpa:
+                    ph, tn = refine_ph(ph)
+                    phones.append(ph)
+                    tones.append(tn)
+                else:
+                    phones.append(ph)
+                    tones.append(0)
+    # todo: implement word2ph
+    word2ph = [1 for i in phones]
+
+    phones = [post_replace_ph(i) for i in phones]
+    return phones, tones, word2ph
+
+if __name__ == "__main__":
+    # print(get_dict())
+    # print(eng_word_to_phoneme("hello"))
+    print(g2p("In this paper, we propose 1 DSPGAN, a GAN-based universal vocoder."))
+    # all_phones = set()
+    # for k, syllables in eng_dict.items():
+    #     for group in syllables:
+    #         for ph in group:
+    #             all_phones.add(ph)
+    # print(all_phones)

text/japanese.py

@@ -0,0 +1,104 @@
+# modified from https://github.com/CjangCjengh/vits/blob/main/text/japanese.py
+import re
+import sys
+
+import pyopenjtalk
+
+from text import symbols
+
+# Regular expression matching Japanese without punctuation marks:
+_japanese_characters = re.compile(
+    r'[A-Za-z\d\u3005\u3040-\u30ff\u4e00-\u9fff\uff11-\uff19\uff21-\uff3a\uff41-\uff5a\uff66-\uff9d]')
+
+# Regular expression matching non-Japanese characters or punctuation marks:
+_japanese_marks = re.compile(
+    r'[^A-Za-z\d\u3005\u3040-\u30ff\u4e00-\u9fff\uff11-\uff19\uff21-\uff3a\uff41-\uff5a\uff66-\uff9d]')
+
+# List of (symbol, Japanese) pairs for marks:
+_symbols_to_japanese = [(re.compile('%s' % x[0]), x[1]) for x in [
+    ('％', 'パーセント')
+]]
+
+
+# List of (consonant, sokuon) pairs:
+_real_sokuon = [(re.compile('%s' % x[0]), x[1]) for x in [
+    (r'Q([↑↓]*[kg])', r'k#\1'),
+    (r'Q([↑↓]*[tdjʧ])', r't#\1'),
+    (r'Q([↑↓]*[sʃ])', r's\1'),
+    (r'Q([↑↓]*[pb])', r'p#\1')
+]]
+
+# List of (consonant, hatsuon) pairs:
+_real_hatsuon = [(re.compile('%s' % x[0]), x[1]) for x in [
+    (r'N([↑↓]*[pbm])', r'm\1'),
+    (r'N([↑↓]*[ʧʥj])', r'n^\1'),
+    (r'N([↑↓]*[tdn])', r'n\1'),
+    (r'N([↑↓]*[kg])', r'ŋ\1')
+]]
+
+
+
+def post_replace_ph(ph):
+    rep_map = {
+        '：': ',',
+        '；': ',',
+        '，': ',',
+        '。': '.',
+        '！': '!',
+        '？': '?',
+        '\n': '.',
+        "·": ",",
+        '、': ",",
+        '...': '…',
+        'v': "V"
+    }
+    if ph in rep_map.keys():
+        ph = rep_map[ph]
+    if ph in symbols:
+        return ph
+    if ph not in symbols:
+        ph = 'UNK'
+    return ph
+
+def symbols_to_japanese(text):
+    for regex, replacement in _symbols_to_japanese:
+        text = re.sub(regex, replacement, text)
+    return text
+
+
+def preprocess_jap(text):
+    '''Reference https://r9y9.github.io/ttslearn/latest/notebooks/ch10_Recipe-Tacotron.html'''
+    text = symbols_to_japanese(text)
+    sentences = re.split(_japanese_marks, text)
+    marks = re.findall(_japanese_marks, text)
+    text = []
+    for i, sentence in enumerate(sentences):
+        if re.match(_japanese_characters, sentence):
+            p = pyopenjtalk.g2p(sentence)
+            text += p.split(" ")
+
+        if i < len(marks):
+            text += [marks[i].replace(' ', '')]
+    return text
+
+def text_normalize(text):
+    # todo: jap text normalize
+    return text
+
+def g2p(norm_text):
+    phones = preprocess_jap(norm_text)
+    phones = [post_replace_ph(i) for i in phones]
+    # todo: implement tones and word2ph
+    tones = [0 for i in phones]
+    word2ph = [1 for i in phones]
+    return phones, tones, word2ph
+
+
+if __name__ == '__main__':
+    for line in open("../../../Downloads/transcript_utf8.txt").readlines():
+        text = line.split(":")[1]
+        phones, tones, word2ph = g2p(text)
+        for p in phones:
+            if p == "z":
+                print(text, phones)
+                sys.exit(0)

text/opencpop-strict.txt

@@ -0,0 +1,429 @@
+a	AA a
+ai	AA ai
+an	AA an
+ang	AA ang
+ao	AA ao
+ba	b a
+bai	b ai
+ban	b an
+bang	b ang
+bao	b ao
+bei	b ei
+ben	b en
+beng	b eng
+bi	b i
+bian	b ian
+biao	b iao
+bie	b ie
+bin	b in
+bing	b ing
+bo	b o
+bu	b u
+ca	c a
+cai	c ai
+can	c an
+cang	c ang
+cao	c ao
+ce	c e
+cei	c ei
+cen	c en
+ceng	c eng
+cha	ch a
+chai	ch ai
+chan	ch an
+chang	ch ang
+chao	ch ao
+che	ch e
+chen	ch en
+cheng	ch eng
+chi	ch ir
+chong	ch ong
+chou	ch ou
+chu	ch u
+chua	ch ua
+chuai	ch uai
+chuan	ch uan
+chuang	ch uang
+chui	ch ui
+chun	ch un
+chuo	ch uo
+ci	c i0
+cong	c ong
+cou	c ou
+cu	c u
+cuan	c uan
+cui	c ui
+cun	c un
+cuo	c uo
+da	d a
+dai	d ai
+dan	d an
+dang	d ang
+dao	d ao
+de	d e
+dei	d ei
+den	d en
+deng	d eng
+di	d i
+dia	d ia
+dian	d ian
+diao	d iao
+die	d ie
+ding	d ing
+diu	d iu
+dong	d ong
+dou	d ou
+du	d u
+duan	d uan
+dui	d ui
+dun	d un
+duo	d uo
+e	EE e
+ei	EE ei
+en	EE en
+eng	EE eng
+er	EE er
+fa	f a
+fan	f an
+fang	f ang
+fei	f ei
+fen	f en
+feng	f eng
+fo	f o
+fou	f ou
+fu	f u
+ga	g a
+gai	g ai
+gan	g an
+gang	g ang
+gao	g ao
+ge	g e
+gei	g ei
+gen	g en
+geng	g eng
+gong	g ong
+gou	g ou
+gu	g u
+gua	g ua
+guai	g uai
+guan	g uan
+guang	g uang
+gui	g ui
+gun	g un
+guo	g uo
+ha	h a
+hai	h ai
+han	h an
+hang	h ang
+hao	h ao
+he	h e
+hei	h ei
+hen	h en
+heng	h eng
+hong	h ong
+hou	h ou
+hu	h u
+hua	h ua
+huai	h uai
+huan	h uan
+huang	h uang
+hui	h ui
+hun	h un
+huo	h uo
+ji	j i
+jia	j ia
+jian	j ian
+jiang	j iang
+jiao	j iao
+jie	j ie
+jin	j in
+jing	j ing
+jiong	j iong
+jiu	j iu
+ju	j v
+jv	j v
+juan	j van
+jvan	j van
+jue	j ve
+jve	j ve
+jun	j vn
+jvn	j vn
+ka	k a
+kai	k ai
+kan	k an
+kang	k ang
+kao	k ao
+ke	k e
+kei	k ei
+ken	k en
+keng	k eng
+kong	k ong
+kou	k ou
+ku	k u
+kua	k ua
+kuai	k uai
+kuan	k uan
+kuang	k uang
+kui	k ui
+kun	k un
+kuo	k uo
+la	l a
+lai	l ai
+lan	l an
+lang	l ang
+lao	l ao
+le	l e
+lei	l ei
+leng	l eng
+li	l i
+lia	l ia
+lian	l ian
+liang	l iang
+liao	l iao
+lie	l ie
+lin	l in
+ling	l ing
+liu	l iu
+lo	l o
+long	l ong
+lou	l ou
+lu	l u
+luan	l uan
+lun	l un
+luo	l uo
+lv	l v
+lve	l ve
+ma	m a
+mai	m ai
+man	m an
+mang	m ang
+mao	m ao
+me	m e
+mei	m ei
+men	m en
+meng	m eng
+mi	m i
+mian	m ian
+miao	m iao
+mie	m ie
+min	m in
+ming	m ing
+miu	m iu
+mo	m o
+mou	m ou
+mu	m u
+na	n a
+nai	n ai
+nan	n an
+nang	n ang
+nao	n ao
+ne	n e
+nei	n ei
+nen	n en
+neng	n eng
+ni	n i
+nian	n ian
+niang	n iang
+niao	n iao
+nie	n ie
+nin	n in
+ning	n ing
+niu	n iu
+nong	n ong
+nou	n ou
+nu	n u
+nuan	n uan
+nun	n un
+nuo	n uo
+nv	n v
+nve	n ve
+o	OO o
+ou	OO ou
+pa	p a
+pai	p ai
+pan	p an
+pang	p ang
+pao	p ao
+pei	p ei
+pen	p en
+peng	p eng
+pi	p i
+pian	p ian
+piao	p iao
+pie	p ie
+pin	p in
+ping	p ing
+po	p o
+pou	p ou
+pu	p u
+qi	q i
+qia	q ia
+qian	q ian
+qiang	q iang
+qiao	q iao
+qie	q ie
+qin	q in
+qing	q ing
+qiong	q iong
+qiu	q iu
+qu	q v
+qv	q v
+quan	q van
+qvan	q van
+que	q ve
+qve	q ve
+qun	q vn
+qvn	q vn
+ran	r an
+rang	r ang
+rao	r ao
+re	r e
+ren	r en
+reng	r eng
+ri	r ir
+rong	r ong
+rou	r ou
+ru	r u
+rua	r ua
+ruan	r uan
+rui	r ui
+run	r un
+ruo	r uo
+sa	s a
+sai	s ai
+san	s an
+sang	s ang
+sao	s ao
+se	s e
+sen	s en
+seng	s eng
+sha	sh a
+shai	sh ai
+shan	sh an
+shang	sh ang
+shao	sh ao
+she	sh e
+shei	sh ei
+shen	sh en
+sheng	sh eng
+shi	sh ir
+shou	sh ou
+shu	sh u
+shua	sh ua
+shuai	sh uai
+shuan	sh uan
+shuang	sh uang
+shui	sh ui
+shun	sh un
+shuo	sh uo
+si	s i0
+song	s ong
+sou	s ou
+su	s u
+suan	s uan
+sui	s ui
+sun	s un
+suo	s uo
+ta	t a
+tai	t ai
+tan	t an
+tang	t ang
+tao	t ao
+te	t e
+tei	t ei
+teng	t eng
+ti	t i
+tian	t ian
+tiao	t iao
+tie	t ie
+ting	t ing
+tong	t ong
+tou	t ou
+tu	t u
+tuan	t uan
+tui	t ui
+tun	t un
+tuo	t uo
+wa	w a
+wai	w ai
+wan	w an
+wang	w ang
+wei	w ei
+wen	w en
+weng	w eng
+wo	w o
+wu	w u
+xi	x i
+xia	x ia
+xian	x ian
+xiang	x iang
+xiao	x iao
+xie	x ie
+xin	x in
+xing	x ing
+xiong	x iong
+xiu	x iu
+xu	x v
+xv	x v
+xuan	x van
+xvan	x van
+xue	x ve
+xve	x ve
+xun	x vn
+xvn	x vn
+ya	y a
+yan	y En
+yang	y ang
+yao	y ao
+ye	y E
+yi	y i
+yin	y in
+ying	y ing
+yo	y o
+yong	y ong
+you	y ou
+yu	y v
+yv	y v
+yuan	y van
+yvan	y van
+yue	y ve
+yve	y ve
+yun	y vn
+yvn	y vn
+za	z a
+zai	z ai
+zan	z an
+zang	z ang
+zao	z ao
+ze	z e
+zei	z ei
+zen	z en
+zeng	z eng
+zha	zh a
+zhai	zh ai
+zhan	zh an
+zhang	zh ang
+zhao	zh ao
+zhe	zh e
+zhei	zh ei
+zhen	zh en
+zheng	zh eng
+zhi	zh ir
+zhong	zh ong
+zhou	zh ou
+zhu	zh u
+zhua	zh ua
+zhuai	zh uai
+zhuan	zh uan
+zhuang	zh uang
+zhui	zh ui
+zhun	zh un
+zhuo	zh uo
+zi	z i0
+zong	z ong
+zou	z ou
+zu	z u
+zuan	z uan
+zui	z ui
+zun	z un
+zuo	z uo

text/symbols.py

@@ -0,0 +1,51 @@
+punctuation = ['!', '?', '…', ",", ".", "'", '-']
+pu_symbols = punctuation + ["SP", "UNK"]
+pad = '_'
+
+# chinese
+zh_symbols = ['E', 'En', 'a', 'ai', 'an', 'ang', 'ao', 'b', 'c', 'ch', 'd', 'e', 'ei', 'en', 'eng', 'er', 'f', 'g', 'h',
+       'i', 'i0', 'ia', 'ian', 'iang', 'iao', 'ie', 'in', 'ing', 'iong', 'ir', 'iu', 'j', 'k', 'l', 'm', 'n', 'o',
+       'ong',
+       'ou', 'p', 'q', 'r', 's', 'sh', 't', 'u', 'ua', 'uai', 'uan', 'uang', 'ui', 'un', 'uo', 'v', 'van', 've', 'vn',
+       'w', 'x', 'y', 'z', 'zh',
+        "AA", "EE", "OO"]
+num_zh_tones = 6
+
+# japanese
+ja_symbols = ['I', 'N', 'U', 'a', 'b', 'by', 'ch', 'cl', 'd', 'dy', 'e', 'f', 'g', 'gy', 'h', 'hy', 'i', 'j', 'k', 'ky',
+              'm', 'my', 'n', 'ny', 'o', 'p', 'py', 'r', 'ry', 's', 'sh', 't', 'ts', 'u', 'V', 'w', 'y', 'z']
+num_ja_tones = 1
+
+# English
+en_symbols = ['aa', 'ae', 'ah', 'ao', 'aw', 'ay', 'b', 'ch', 'd', 'dh', 'eh', 'er', 'ey', 'f', 'g', 'hh', 'ih', 'iy',
+              'jh', 'k', 'l', 'm', 'n', 'ng', 'ow', 'oy', 'p', 'r', 's',
+              'sh', 't', 'th', 'uh', 'uw', 'V', 'w', 'y', 'z', 'zh']
+num_en_tones = 4
+
+# combine all symbols
+normal_symbols = sorted(set(zh_symbols + ja_symbols + en_symbols))
+symbols = [pad] + normal_symbols + pu_symbols
+sil_phonemes_ids = [symbols.index(i) for i in pu_symbols]
+
+# combine all tones
+num_tones = num_zh_tones + num_ja_tones + num_en_tones
+
+# language maps
+language_id_map = {
+    'ZH': 0,
+    "JA": 1,
+    "EN": 2
+}
+num_languages = len(language_id_map.keys())
+
+language_tone_start_map = {
+    'ZH': 0,
+    "JA": num_zh_tones,
+    "EN": num_zh_tones + num_ja_tones
+}
+
+if __name__ == '__main__':
+    a = set(zh_symbols)
+    b = set(en_symbols)
+    print(sorted(a&b))
+

text/tone_sandhi.py

@@ -0,0 +1,351 @@
+# Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+from typing import List
+from typing import Tuple
+
+import jieba
+from pypinyin import lazy_pinyin
+from pypinyin import Style
+
+
+class ToneSandhi():
+    def __init__(self):
+        self.must_neural_tone_words = {
+            '麻烦', '麻利', '鸳鸯', '高粱', '骨头', '骆驼', '马虎', '首饰', '馒头', '馄饨', '风筝',
+            '难为', '队伍', '阔气', '闺女', '门道', '锄头', '铺盖', '铃铛', '铁匠', '钥匙', '里脊',
+            '里头', '部分', '那么', '道士', '造化', '迷糊', '连累', '这么', '这个', '运气', '过去',
+            '软和', '转悠', '踏实', '跳蚤', '跟头', '趔趄', '财主', '豆腐', '讲究', '记性', '记号',
+            '认识', '规矩', '见识', '裁缝', '补丁', '衣裳', '衣服', '衙门', '街坊', '行李', '行当',
+            '蛤蟆', '蘑菇', '薄荷', '葫芦', '葡萄', '萝卜', '荸荠', '苗条', '苗头', '苍蝇', '芝麻',
+            '舒服', '舒坦', '舌头', '自在', '膏药', '脾气', '脑袋', '脊梁', '能耐', '胳膊', '胭脂',
+            '胡萝', '胡琴', '胡同', '聪明', '耽误', '耽搁', '耷拉', '耳朵', '老爷', '老实', '老婆',
+            '老头', '老太', '翻腾', '罗嗦', '罐头', '编辑', '结实', '红火', '累赘', '糨糊', '糊涂',
+            '精神', '粮食', '簸箕', '篱笆', '算计', '算盘', '答应', '笤帚', '笑语', '笑话', '窟窿',
+            '窝囊', '窗户', '稳当', '稀罕', '称呼', '秧歌', '秀气', '秀才', '福气', '祖宗', '砚台',
+            '码头', '石榴', '石头', '石匠', '知识', '眼睛', '眯缝', '眨巴', '眉毛', '相声', '盘算',
+            '白净', '痢疾', '痛快', '疟疾', '疙瘩', '疏忽', '畜生', '生意', '甘蔗', '琵琶', '琢磨',
+            '琉璃', '玻璃', '玫瑰', '玄乎', '狐狸', '状元', '特务', '牲口', '牙碜', '牌楼', '爽快',
+            '爱人', '热闹', '烧饼', '烟筒', '烂糊', '点心', '炊帚', '灯笼', '火候', '漂亮', '滑溜',
+            '溜达', '温和', '清楚', '消息', '浪头', '活泼', '比方', '正经', '欺负', '模糊', '槟榔',
+            '棺材', '棒槌', '棉花', '核桃', '栅栏', '柴火', '架势', '枕头', '枇杷', '机灵', '本事',
+            '木头', '木匠', '朋友', '月饼', '月亮', '暖和', '明白', '时候', '新鲜', '故事', '收拾',
+            '收成', '提防', '挖苦', '挑剔', '指甲', '指头', '拾掇', '拳头', '拨弄', '招牌', '招呼',
+            '抬举', '护士', '折腾', '扫帚', '打量', '打算', '打点', '打扮', '打听', '打发', '扎实',
+            '扁担', '戒指', '懒得', '意识', '意思', '情形', '悟性', '怪物', '思量', '怎么', '念头',
+            '念叨', '快活', '忙活', '志气', '心思', '得罪', '张罗', '弟兄', '开通', '应酬', '庄稼',
+            '干事', '帮手', '帐篷', '希罕', '师父', '师傅', '巴结', '巴掌', '差事', '工夫', '岁数',
+            '屁股', '尾巴', '少爷', '小气', '小伙', '将就', '对头', '对付', '寡妇', '家伙', '客气',
+            '实在', '官司', '学问', '学生', '字号', '嫁妆', '媳妇', '媒人', '婆家', '娘家', '委屈',
+            '姑娘', '姐夫', '妯娌', '妥当', '妖精', '奴才', '女婿', '头发', '太阳', '大爷', '大方',
+            '大意', '大夫', '多少', '多么', '外甥', '壮实', '地道', '地方', '在乎', '困难', '嘴巴',
+            '嘱咐', '嘟囔', '嘀咕', '喜欢', '喇嘛', '喇叭', '商量', '唾沫', '哑巴', '哈欠', '哆嗦',
+            '咳嗽', '和尚', '告诉', '告示', '含糊', '吓唬', '后头', '名字', '名堂', '合同', '吆喝',
+            '叫唤', '口袋', '厚道', '厉害', '千斤', '包袱', '包涵', '匀称', '勤快', '动静', '动弹',
+            '功夫', '力气', '前头', '刺猬', '刺激', '别扭', '利落', '利索', '利害', '分析', '出息',
+            '凑合', '凉快', '冷战', '冤枉', '冒失', '养活', '关系', '先生', '兄弟', '便宜', '使唤',
+            '佩服', '作坊', '体面', '位置', '似的', '伙计', '休息', '什么', '人家', '亲戚', '亲家',
+            '交情', '云彩', '事情', '买卖', '主意', '丫头', '丧气', '两口', '东西', '东家', '世故',
+            '不由', '不在', '下水', '下巴', '上头', '上司', '丈夫', '丈人', '一辈', '那个', '菩萨',
+            '父亲', '母亲', '咕噜', '邋遢', '费用', '冤家', '甜头', '介绍', '荒唐', '大人', '泥鳅',
+            '幸福', '熟悉', '计划', '扑腾', '蜡烛', '姥爷', '照顾', '喉咙', '吉他', '弄堂', '蚂蚱',
+            '凤凰', '拖沓', '寒碜', '糟蹋', '倒腾', '报复', '逻辑', '盘缠', '喽啰', '牢骚', '咖喱',
+            '扫把', '惦记'
+        }
+        self.must_not_neural_tone_words = {
+            "男子", "女子", "分子", "原子", "量子", "莲子", "石子", "瓜子", "电子", "人人", "虎虎"
+        }
+        self.punc = "：，；。？！“”‘’':,;.?!"
+
+    # the meaning of jieba pos tag: https://blog.csdn.net/weixin_44174352/article/details/113731041
+    # e.g.
+    # word: "家里"
+    # pos: "s"
+    # finals: ['ia1', 'i3']
+    def _neural_sandhi(self, word: str, pos: str,
+                       finals: List[str]) -> List[str]:
+
+        # reduplication words for n. and v. e.g. 奶奶, 试试, 旺旺
+        for j, item in enumerate(word):
+            if j - 1 >= 0 and item == word[j - 1] and pos[0] in {
+                    "n", "v", "a"
+            } and word not in self.must_not_neural_tone_words:
+                finals[j] = finals[j][:-1] + "5"
+        ge_idx = word.find("个")
+        if len(word) >= 1 and word[-1] in "吧呢哈啊呐噻嘛吖嗨呐哦哒额滴哩哟喽啰耶喔诶":
+            finals[-1] = finals[-1][:-1] + "5"
+        elif len(word) >= 1 and word[-1] in "的地得":
+            finals[-1] = finals[-1][:-1] + "5"
+        # e.g. 走了, 看着, 去过
+        elif len(word) == 1 and word in "了着过" and pos in {"ul", "uz", "ug"}:
+            finals[-1] = finals[-1][:-1] + "5"
+        elif len(word) > 1 and word[-1] in "们子" and pos in {
+                "r", "n"
+        } and word not in self.must_not_neural_tone_words:
+            finals[-1] = finals[-1][:-1] + "5"
+        # e.g. 桌上, 地下, 家里
+        elif len(word) > 1 and word[-1] in "上下里" and pos in {"s", "l", "f"}:
+            finals[-1] = finals[-1][:-1] + "5"
+        # e.g. 上来, 下去
+        elif len(word) > 1 and word[-1] in "来去" and word[-2] in "上下进出回过起开":
+            finals[-1] = finals[-1][:-1] + "5"
+        # 个做量词
+        elif (ge_idx >= 1 and
+              (word[ge_idx - 1].isnumeric() or
+               word[ge_idx - 1] in "几有两半多各整每做是")) or word == '个':
+            finals[ge_idx] = finals[ge_idx][:-1] + "5"
+        else:
+            if word in self.must_neural_tone_words or word[
+                    -2:] in self.must_neural_tone_words:
+                finals[-1] = finals[-1][:-1] + "5"
+
+        word_list = self._split_word(word)
+        finals_list = [finals[:len(word_list[0])], finals[len(word_list[0]):]]
+        for i, word in enumerate(word_list):
+            # conventional neural in Chinese
+            if word in self.must_neural_tone_words or word[
+                    -2:] in self.must_neural_tone_words:
+                finals_list[i][-1] = finals_list[i][-1][:-1] + "5"
+        finals = sum(finals_list, [])
+        return finals
+
+    def _bu_sandhi(self, word: str, finals: List[str]) -> List[str]:
+        # e.g. 看不懂
+        if len(word) == 3 and word[1] == "不":
+            finals[1] = finals[1][:-1] + "5"
+        else:
+            for i, char in enumerate(word):
+                # "不" before tone4 should be bu2, e.g. 不怕
+                if char == "不" and i + 1 < len(word) and finals[i +
+                                                                1][-1] == "4":
+                    finals[i] = finals[i][:-1] + "2"
+        return finals
+
+    def _yi_sandhi(self, word: str, finals: List[str]) -> List[str]:
+        # "一" in number sequences, e.g. 一零零, 二一零
+        if word.find("一") != -1 and all(
+            [item.isnumeric() for item in word if item != "一"]):
+            return finals
+        # "一" between reduplication words shold be yi5, e.g. 看一看
+        elif len(word) == 3 and word[1] == "一" and word[0] == word[-1]:
+            finals[1] = finals[1][:-1] + "5"
+        # when "一" is ordinal word, it should be yi1
+        elif word.startswith("第一"):
+            finals[1] = finals[1][:-1] + "1"
+        else:
+            for i, char in enumerate(word):
+                if char == "一" and i + 1 < len(word):
+                    # "一" before tone4 should be yi2, e.g. 一段
+                    if finals[i + 1][-1] == "4":
+                        finals[i] = finals[i][:-1] + "2"
+                    # "一" before non-tone4 should be yi4, e.g. 一天
+                    else:
+                        # "一" 后面如果是标点，还读一声
+                        if word[i + 1] not in self.punc:
+                            finals[i] = finals[i][:-1] + "4"
+        return finals
+
+    def _split_word(self, word: str) -> List[str]:
+        word_list = jieba.cut_for_search(word)
+        word_list = sorted(word_list, key=lambda i: len(i), reverse=False)
+        first_subword = word_list[0]
+        first_begin_idx = word.find(first_subword)
+        if first_begin_idx == 0:
+            second_subword = word[len(first_subword):]
+            new_word_list = [first_subword, second_subword]
+        else:
+            second_subword = word[:-len(first_subword)]
+            new_word_list = [second_subword, first_subword]
+        return new_word_list
+
+    def _three_sandhi(self, word: str, finals: List[str]) -> List[str]:
+        if len(word) == 2 and self._all_tone_three(finals):
+            finals[0] = finals[0][:-1] + "2"
+        elif len(word) == 3:
+            word_list = self._split_word(word)
+            if self._all_tone_three(finals):
+                #  disyllabic + monosyllabic, e.g. 蒙古/包
+                if len(word_list[0]) == 2:
+                    finals[0] = finals[0][:-1] + "2"
+                    finals[1] = finals[1][:-1] + "2"
+                #  monosyllabic + disyllabic, e.g. 纸/老虎
+                elif len(word_list[0]) == 1:
+                    finals[1] = finals[1][:-1] + "2"
+            else:
+                finals_list = [
+                    finals[:len(word_list[0])], finals[len(word_list[0]):]
+                ]
+                if len(finals_list) == 2:
+                    for i, sub in enumerate(finals_list):
+                        # e.g. 所有/人
+                        if self._all_tone_three(sub) and len(sub) == 2:
+                            finals_list[i][0] = finals_list[i][0][:-1] + "2"
+                        # e.g. 好/喜欢
+                        elif i == 1 and not self._all_tone_three(sub) and finals_list[i][0][-1] == "3" and \
+                                finals_list[0][-1][-1] == "3":
+
+                            finals_list[0][-1] = finals_list[0][-1][:-1] + "2"
+                        finals = sum(finals_list, [])
+        # split idiom into two words who's length is 2
+        elif len(word) == 4:
+            finals_list = [finals[:2], finals[2:]]
+            finals = []
+            for sub in finals_list:
+                if self._all_tone_three(sub):
+                    sub[0] = sub[0][:-1] + "2"
+                finals += sub
+
+        return finals
+
+    def _all_tone_three(self, finals: List[str]) -> bool:
+        return all(x[-1] == "3" for x in finals)
+
+    # merge "不" and the word behind it
+    # if don't merge, "不" sometimes appears alone according to jieba, which may occur sandhi error
+    def _merge_bu(self, seg: List[Tuple[str, str]]) -> List[Tuple[str, str]]:
+        new_seg = []
+        last_word = ""
+        for word, pos in seg:
+            if last_word == "不":
+                word = last_word + word
+            if word != "不":
+                new_seg.append((word, pos))
+            last_word = word[:]
+        if last_word == "不":
+            new_seg.append((last_word, 'd'))
+            last_word = ""
+        return new_seg
+
+    # function 1: merge "一" and reduplication words in it's left and right, e.g. "听","一","听" ->"听一听"
+    # function 2: merge single  "一" and the word behind it
+    # if don't merge, "一" sometimes appears alone according to jieba, which may occur sandhi error
+    # e.g.
+    # input seg: [('听', 'v'), ('一', 'm'), ('听', 'v')]
+    # output seg: [['听一听', 'v']]
+    def _merge_yi(self, seg: List[Tuple[str, str]]) -> List[Tuple[str, str]]:
+        new_seg = []
+        # function 1
+        for i, (word, pos) in enumerate(seg):
+            if i - 1 >= 0 and word == "一" and i + 1 < len(seg) and seg[i - 1][
+                    0] == seg[i + 1][0] and seg[i - 1][1] == "v":
+                new_seg[i - 1][0] = new_seg[i - 1][0] + "一" + new_seg[i - 1][0]
+            else:
+                if i - 2 >= 0 and seg[i - 1][0] == "一" and seg[i - 2][
+                        0] == word and pos == "v":
+                    continue
+                else:
+                    new_seg.append([word, pos])
+        seg = new_seg
+        new_seg = []
+        # function 2
+        for i, (word, pos) in enumerate(seg):
+            if new_seg and new_seg[-1][0] == "一":
+                new_seg[-1][0] = new_seg[-1][0] + word
+            else:
+                new_seg.append([word, pos])
+        return new_seg
+
+    # the first and the second words are all_tone_three
+    def _merge_continuous_three_tones(
+            self, seg: List[Tuple[str, str]]) -> List[Tuple[str, str]]:
+        new_seg = []
+        sub_finals_list = [
+            lazy_pinyin(
+                word, neutral_tone_with_five=True, style=Style.FINALS_TONE3)
+            for (word, pos) in seg
+        ]
+        assert len(sub_finals_list) == len(seg)
+        merge_last = [False] * len(seg)
+        for i, (word, pos) in enumerate(seg):
+            if i - 1 >= 0 and self._all_tone_three(
+                    sub_finals_list[i - 1]) and self._all_tone_three(
+                        sub_finals_list[i]) and not merge_last[i - 1]:
+                # if the last word is reduplication, not merge, because reduplication need to be _neural_sandhi
+                if not self._is_reduplication(seg[i - 1][0]) and len(
+                        seg[i - 1][0]) + len(seg[i][0]) <= 3:
+                    new_seg[-1][0] = new_seg[-1][0] + seg[i][0]
+                    merge_last[i] = True
+                else:
+                    new_seg.append([word, pos])
+            else:
+                new_seg.append([word, pos])
+
+        return new_seg
+
+    def _is_reduplication(self, word: str) -> bool:
+        return len(word) == 2 and word[0] == word[1]
+
+    # the last char of first word and the first char of second word is tone_three
+    def _merge_continuous_three_tones_2(
+            self, seg: List[Tuple[str, str]]) -> List[Tuple[str, str]]:
+        new_seg = []
+        sub_finals_list = [
+            lazy_pinyin(
+                word, neutral_tone_with_five=True, style=Style.FINALS_TONE3)
+            for (word, pos) in seg
+        ]
+        assert len(sub_finals_list) == len(seg)
+        merge_last = [False] * len(seg)
+        for i, (word, pos) in enumerate(seg):
+            if i - 1 >= 0 and sub_finals_list[i - 1][-1][-1] == "3" and sub_finals_list[i][0][-1] == "3" and not \
+                    merge_last[i - 1]:
+                # if the last word is reduplication, not merge, because reduplication need to be _neural_sandhi
+                if not self._is_reduplication(seg[i - 1][0]) and len(
+                        seg[i - 1][0]) + len(seg[i][0]) <= 3:
+                    new_seg[-1][0] = new_seg[-1][0] + seg[i][0]
+                    merge_last[i] = True
+                else:
+                    new_seg.append([word, pos])
+            else:
+                new_seg.append([word, pos])
+        return new_seg
+
+    def _merge_er(self, seg: List[Tuple[str, str]]) -> List[Tuple[str, str]]:
+        new_seg = []
+        for i, (word, pos) in enumerate(seg):
+            if i - 1 >= 0 and word == "儿" and seg[i-1][0] != "#":
+                new_seg[-1][0] = new_seg[-1][0] + seg[i][0]
+            else:
+                new_seg.append([word, pos])
+        return new_seg
+
+    def _merge_reduplication(
+            self, seg: List[Tuple[str, str]]) -> List[Tuple[str, str]]:
+        new_seg = []
+        for i, (word, pos) in enumerate(seg):
+            if new_seg and word == new_seg[-1][0]:
+                new_seg[-1][0] = new_seg[-1][0] + seg[i][0]
+            else:
+                new_seg.append([word, pos])
+        return new_seg
+
+    def pre_merge_for_modify(
+            self, seg: List[Tuple[str, str]]) -> List[Tuple[str, str]]:
+        seg = self._merge_bu(seg)
+        try:
+            seg = self._merge_yi(seg)
+        except:
+            print("_merge_yi failed")
+        seg = self._merge_reduplication(seg)
+        seg = self._merge_continuous_three_tones(seg)
+        seg = self._merge_continuous_three_tones_2(seg)
+        seg = self._merge_er(seg)
+        return seg
+
+    def modified_tone(self, word: str, pos: str,
+                      finals: List[str]) -> List[str]:
+        finals = self._bu_sandhi(word, finals)
+        finals = self._yi_sandhi(word, finals)
+        finals = self._neural_sandhi(word, pos, finals)
+        finals = self._three_sandhi(word, finals)
+        return finals

train_ms.py

@@ -0,0 +1,322 @@
+import os
+import json
+import argparse
+import itertools
+import math
+import torch
+from torch import nn, optim
+from torch.nn import functional as F
+from torch.utils.data import DataLoader
+from torch.utils.tensorboard import SummaryWriter
+import torch.multiprocessing as mp
+import torch.distributed as dist
+from torch.nn.parallel import DistributedDataParallel as DDP
+from torch.cuda.amp import autocast, GradScaler
+from tqdm import tqdm
+
+import commons
+import utils
+from data_utils import (
+    TextAudioSpeakerLoader,
+    TextAudioSpeakerCollate,
+    DistributedBucketSampler
+)
+from models import (
+    SynthesizerTrn,
+    MultiPeriodDiscriminator,
+)
+from losses import (
+    generator_loss,
+    discriminator_loss,
+    feature_loss,
+    kl_loss
+)
+from mel_processing import mel_spectrogram_torch, spec_to_mel_torch
+from text.symbols import symbols
+
+torch.backends.cudnn.benchmark = True
+global_step = 0
+
+
+def main():
+    """Assume Single Node Multi GPUs Training Only"""
+    assert torch.cuda.is_available(), "CPU training is not allowed."
+
+    n_gpus = torch.cuda.device_count()
+    os.environ['MASTER_ADDR'] = 'localhost'
+    os.environ['MASTER_PORT'] = '8000'
+
+    hps = utils.get_hparams()
+    mp.spawn(run, nprocs=n_gpus, args=(n_gpus, hps,))
+
+
+def run(rank, n_gpus, hps):
+    global global_step
+    if rank == 0:
+        logger = utils.get_logger(hps.model_dir)
+        logger.info(hps)
+        utils.check_git_hash(hps.model_dir)
+        writer = SummaryWriter(log_dir=hps.model_dir)
+        writer_eval = SummaryWriter(log_dir=os.path.join(hps.model_dir, "eval"))
+
+    dist.init_process_group(backend='nccl', init_method='env://', world_size=n_gpus, rank=rank)
+    torch.manual_seed(hps.train.seed)
+    torch.cuda.set_device(rank)
+
+    train_dataset = TextAudioSpeakerLoader(hps.data.training_files, hps.data)
+    train_sampler = DistributedBucketSampler(
+        train_dataset,
+        hps.train.batch_size,
+        [32, 300, 400, 500, 600, 700, 800, 900, 1000],
+        num_replicas=n_gpus,
+        rank=rank,
+        shuffle=True)
+    collate_fn = TextAudioSpeakerCollate()
+    train_loader = DataLoader(train_dataset, num_workers=4, shuffle=False, pin_memory=True,
+                              collate_fn=collate_fn, batch_sampler=train_sampler, persistent_workers=True)
+    if rank == 0:
+        eval_dataset = TextAudioSpeakerLoader(hps.data.validation_files, hps.data)
+        eval_loader = DataLoader(eval_dataset, num_workers=0, shuffle=False,
+                                 batch_size=1, pin_memory=True,
+                                 drop_last=False, collate_fn=collate_fn)
+
+    net_g = SynthesizerTrn(
+        len(symbols),
+        hps.data.filter_length // 2 + 1,
+        hps.train.segment_size // hps.data.hop_length,
+        n_speakers=hps.data.n_speakers,
+        **hps.model).cuda(rank)
+
+    freeze_enc = getattr(hps.model, "freeze_enc", False)
+    if freeze_enc:
+        print("freeze encoder !!!")
+        for param in net_g.enc_p.parameters():
+            param.requires_grad = False
+
+    net_d = MultiPeriodDiscriminator(hps.model.use_spectral_norm).cuda(rank)
+    optim_g = torch.optim.AdamW(
+        filter(lambda p: p.requires_grad, net_g.parameters()),
+        hps.train.learning_rate,
+        betas=hps.train.betas,
+        eps=hps.train.eps)
+    optim_d = torch.optim.AdamW(
+        net_d.parameters(),
+        hps.train.learning_rate,
+        betas=hps.train.betas,
+        eps=hps.train.eps)
+    net_g = DDP(net_g, device_ids=[rank])
+    net_d = DDP(net_d, device_ids=[rank])
+
+    pretrain_dir = "logs/esd"
+    if pretrain_dir is None:
+        _, _, _, epoch_str = utils.load_checkpoint(utils.latest_checkpoint_path(hps.model_dir, "G_*.pth"), net_g,
+                                                   optim_g, False)
+        _, _, _, epoch_str = utils.load_checkpoint(utils.latest_checkpoint_path(hps.model_dir, "D_*.pth"), net_d,
+                                                   optim_d, False)
+        epoch_str = max(epoch_str, 1)
+        global_step = (epoch_str - 1) * len(train_loader)
+    else:
+        _, _, _, epoch_str = utils.load_checkpoint(utils.latest_checkpoint_path(pretrain_dir, "G_*.pth"), net_g,
+                                                   optim_g, True)
+        _, _, _, epoch_str = utils.load_checkpoint(utils.latest_checkpoint_path(pretrain_dir, "D_*.pth"), net_d,
+                                                   optim_d, True)
+        epoch_str = 1
+        global_step = 0
+
+
+
+    scheduler_g = torch.optim.lr_scheduler.ExponentialLR(optim_g, gamma=hps.train.lr_decay, last_epoch=epoch_str - 2)
+    scheduler_d = torch.optim.lr_scheduler.ExponentialLR(optim_d, gamma=hps.train.lr_decay, last_epoch=epoch_str - 2)
+
+    scaler = GradScaler(enabled=hps.train.fp16_run)
+
+    for epoch in range(epoch_str, hps.train.epochs + 1):
+        if rank == 0:
+            train_and_evaluate(rank, epoch, hps, [net_g, net_d], [optim_g, optim_d], [scheduler_g, scheduler_d], scaler,
+                               [train_loader, eval_loader], logger, [writer, writer_eval])
+        else:
+            train_and_evaluate(rank, epoch, hps, [net_g, net_d], [optim_g, optim_d], [scheduler_g, scheduler_d], scaler,
+                               [train_loader, None], None, None)
+        scheduler_g.step()
+        scheduler_d.step()
+
+
+def train_and_evaluate(rank, epoch, hps, nets, optims, schedulers, scaler, loaders, logger, writers):
+    net_g, net_d = nets
+    optim_g, optim_d = optims
+    scheduler_g, scheduler_d = schedulers
+    train_loader, eval_loader = loaders
+    if writers is not None:
+        writer, writer_eval = writers
+
+    train_loader.batch_sampler.set_epoch(epoch)
+    global global_step
+
+    net_g.train()
+    net_d.train()
+    for batch_idx, (x, x_lengths, spec, spec_lengths, y, y_lengths, speakers, tone, language, bert) in tqdm(enumerate(train_loader)):
+        x, x_lengths = x.cuda(rank, non_blocking=True), x_lengths.cuda(rank, non_blocking=True)
+        spec, spec_lengths = spec.cuda(rank, non_blocking=True), spec_lengths.cuda(rank, non_blocking=True)
+        y, y_lengths = y.cuda(rank, non_blocking=True), y_lengths.cuda(rank, non_blocking=True)
+        speakers = speakers.cuda(rank, non_blocking=True)
+        tone = tone.cuda(rank, non_blocking=True)
+        language = language.cuda(rank, non_blocking=True)
+        bert = bert.cuda(rank, non_blocking=True)
+
+        with autocast(enabled=hps.train.fp16_run):
+            y_hat, l_length, attn, ids_slice, x_mask, z_mask, \
+                (z, z_p, m_p, logs_p, m_q, logs_q) = net_g(x, x_lengths, spec, spec_lengths, speakers, tone, language, bert)
+
+            mel = spec_to_mel_torch(
+                spec,
+                hps.data.filter_length,
+                hps.data.n_mel_channels,
+                hps.data.sampling_rate,
+                hps.data.mel_fmin,
+                hps.data.mel_fmax)
+            y_mel = commons.slice_segments(mel, ids_slice, hps.train.segment_size // hps.data.hop_length)
+            y_hat_mel = mel_spectrogram_torch(
+                y_hat.squeeze(1),
+                hps.data.filter_length,
+                hps.data.n_mel_channels,
+                hps.data.sampling_rate,
+                hps.data.hop_length,
+                hps.data.win_length,
+                hps.data.mel_fmin,
+                hps.data.mel_fmax
+            )
+
+            y = commons.slice_segments(y, ids_slice * hps.data.hop_length, hps.train.segment_size)  # slice
+
+            # Discriminator
+            y_d_hat_r, y_d_hat_g, _, _ = net_d(y, y_hat.detach())
+            with autocast(enabled=False):
+                loss_disc, losses_disc_r, losses_disc_g = discriminator_loss(y_d_hat_r, y_d_hat_g)
+                loss_disc_all = loss_disc
+        optim_d.zero_grad()
+        scaler.scale(loss_disc_all).backward()
+        scaler.unscale_(optim_d)
+        grad_norm_d = commons.clip_grad_value_(net_d.parameters(), None)
+        scaler.step(optim_d)
+
+        with autocast(enabled=hps.train.fp16_run):
+            # Generator
+            y_d_hat_r, y_d_hat_g, fmap_r, fmap_g = net_d(y, y_hat)
+            with autocast(enabled=False):
+                loss_dur = torch.sum(l_length.float())
+                loss_mel = F.l1_loss(y_mel, y_hat_mel) * hps.train.c_mel
+                loss_kl = kl_loss(z_p, logs_q, m_p, logs_p, z_mask) * hps.train.c_kl
+
+                loss_fm = feature_loss(fmap_r, fmap_g)
+                loss_gen, losses_gen = generator_loss(y_d_hat_g)
+                loss_gen_all = loss_gen + loss_fm + loss_mel + loss_dur + loss_kl
+        optim_g.zero_grad()
+        scaler.scale(loss_gen_all).backward()
+        scaler.unscale_(optim_g)
+        grad_norm_g = commons.clip_grad_value_(net_g.parameters(), None)
+        scaler.step(optim_g)
+        scaler.update()
+
+        if rank == 0:
+            if global_step % hps.train.log_interval == 0:
+                lr = optim_g.param_groups[0]['lr']
+                losses = [loss_disc, loss_gen, loss_fm, loss_mel, loss_dur, loss_kl]
+                logger.info('Train Epoch: {} [{:.0f}%]'.format(
+                    epoch,
+                    100. * batch_idx / len(train_loader)))
+                logger.info([x.item() for x in losses] + [global_step, lr])
+
+                scalar_dict = {"loss/g/total": loss_gen_all, "loss/d/total": loss_disc_all, "learning_rate": lr,
+                               "grad_norm_d": grad_norm_d, "grad_norm_g": grad_norm_g}
+                scalar_dict.update(
+                    {"loss/g/fm": loss_fm, "loss/g/mel": loss_mel, "loss/g/dur": loss_dur, "loss/g/kl": loss_kl})
+
+                scalar_dict.update({"loss/g/{}".format(i): v for i, v in enumerate(losses_gen)})
+                scalar_dict.update({"loss/d_r/{}".format(i): v for i, v in enumerate(losses_disc_r)})
+                scalar_dict.update({"loss/d_g/{}".format(i): v for i, v in enumerate(losses_disc_g)})
+                image_dict = {
+                    "slice/mel_org": utils.plot_spectrogram_to_numpy(y_mel[0].data.cpu().numpy()),
+                    "slice/mel_gen": utils.plot_spectrogram_to_numpy(y_hat_mel[0].data.cpu().numpy()),
+                    "all/mel": utils.plot_spectrogram_to_numpy(mel[0].data.cpu().numpy()),
+                    "all/attn": utils.plot_alignment_to_numpy(attn[0, 0].data.cpu().numpy())
+                }
+                utils.summarize(
+                    writer=writer,
+                    global_step=global_step,
+                    images=image_dict,
+                    scalars=scalar_dict)
+
+            if global_step % hps.train.eval_interval == 0:
+                evaluate(hps, net_g, eval_loader, writer_eval)
+                utils.save_checkpoint(net_g, optim_g, hps.train.learning_rate, epoch,
+                                      os.path.join(hps.model_dir, "G_{}.pth".format(global_step)))
+                utils.save_checkpoint(net_d, optim_d, hps.train.learning_rate, epoch,
+                                      os.path.join(hps.model_dir, "D_{}.pth".format(global_step)))
+                keep_ckpts = getattr(hps.train, 'keep_ckpts', 3)
+                if keep_ckpts > 0:
+                    utils.clean_checkpoints(path_to_models=hps.model_dir, n_ckpts_to_keep=keep_ckpts, sort_by_time=True)
+
+
+        global_step += 1
+
+    if rank == 0:
+        logger.info('====> Epoch: {}'.format(epoch))
+
+
+
+def evaluate(hps, generator, eval_loader, writer_eval):
+    generator.eval()
+    image_dict = {}
+    audio_dict = {}
+    print("Evaluating ...")
+    with torch.no_grad():
+        for batch_idx, (x, x_lengths, spec, spec_lengths, y, y_lengths, speakers, tone, language, bert) in enumerate(eval_loader):
+            print(111)
+            x, x_lengths = x.cuda(), x_lengths.cuda()
+            spec, spec_lengths = spec.cuda(), spec_lengths.cuda()
+            y, y_lengths = y.cuda(), y_lengths.cuda()
+            speakers = speakers.cuda()
+            bert = bert.cuda()
+            tone = tone.cuda()
+            language = language.cuda()
+            for use_sdp in [False, True]:
+                y_hat, attn, mask, *_ = generator.module.infer(x, x_lengths, speakers, tone, language, bert, y=spec, max_len=1000, sdp_ratio=0.0 if not use_sdp else 1.0)
+                y_hat_lengths = mask.sum([1, 2]).long() * hps.data.hop_length
+
+                mel = spec_to_mel_torch(
+                    spec,
+                    hps.data.filter_length,
+                    hps.data.n_mel_channels,
+                    hps.data.sampling_rate,
+                    hps.data.mel_fmin,
+                    hps.data.mel_fmax)
+                y_hat_mel = mel_spectrogram_torch(
+                    y_hat.squeeze(1).float(),
+                    hps.data.filter_length,
+                    hps.data.n_mel_channels,
+                    hps.data.sampling_rate,
+                    hps.data.hop_length,
+                    hps.data.win_length,
+                    hps.data.mel_fmin,
+                    hps.data.mel_fmax
+                )
+                image_dict.update({
+                    f"gen/mel_{batch_idx}": utils.plot_spectrogram_to_numpy(y_hat_mel[0].cpu().numpy())
+                })
+                audio_dict.update({
+                    f"gen/audio_{batch_idx}_{use_sdp}": y_hat[0, :, :y_hat_lengths[0]]
+                })
+                image_dict.update({f"gt/mel_{batch_idx}": utils.plot_spectrogram_to_numpy(mel[0].cpu().numpy())})
+                audio_dict.update({f"gt/audio_{batch_idx}": y[0, :, :y_lengths[0]]})
+
+    utils.summarize(
+        writer=writer_eval,
+        global_step=global_step,
+        images=image_dict,
+        audios=audio_dict,
+        audio_sampling_rate=hps.data.sampling_rate
+    )
+    generator.train()
+
+if __name__ == "__main__":
+    main()

transforms.py

@@ -0,0 +1,193 @@
+import torch
+from torch.nn import functional as F
+
+import numpy as np
+
+
+DEFAULT_MIN_BIN_WIDTH = 1e-3
+DEFAULT_MIN_BIN_HEIGHT = 1e-3
+DEFAULT_MIN_DERIVATIVE = 1e-3
+
+
+def piecewise_rational_quadratic_transform(inputs, 
+                                           unnormalized_widths,
+                                           unnormalized_heights,
+                                           unnormalized_derivatives,
+                                           inverse=False,
+                                           tails=None, 
+                                           tail_bound=1.,
+                                           min_bin_width=DEFAULT_MIN_BIN_WIDTH,
+                                           min_bin_height=DEFAULT_MIN_BIN_HEIGHT,
+                                           min_derivative=DEFAULT_MIN_DERIVATIVE):
+
+    if tails is None:
+        spline_fn = rational_quadratic_spline
+        spline_kwargs = {}
+    else:
+        spline_fn = unconstrained_rational_quadratic_spline
+        spline_kwargs = {
+            'tails': tails,
+            'tail_bound': tail_bound
+        }
+
+    outputs, logabsdet = spline_fn(
+            inputs=inputs,
+            unnormalized_widths=unnormalized_widths,
+            unnormalized_heights=unnormalized_heights,
+            unnormalized_derivatives=unnormalized_derivatives,
+            inverse=inverse,
+            min_bin_width=min_bin_width,
+            min_bin_height=min_bin_height,
+            min_derivative=min_derivative,
+            **spline_kwargs
+    )
+    return outputs, logabsdet
+
+
+def searchsorted(bin_locations, inputs, eps=1e-6):
+    bin_locations[..., -1] += eps
+    return torch.sum(
+        inputs[..., None] >= bin_locations,
+        dim=-1
+    ) - 1
+
+
+def unconstrained_rational_quadratic_spline(inputs,
+                                            unnormalized_widths,
+                                            unnormalized_heights,
+                                            unnormalized_derivatives,
+                                            inverse=False,
+                                            tails='linear',
+                                            tail_bound=1.,
+                                            min_bin_width=DEFAULT_MIN_BIN_WIDTH,
+                                            min_bin_height=DEFAULT_MIN_BIN_HEIGHT,
+                                            min_derivative=DEFAULT_MIN_DERIVATIVE):
+    inside_interval_mask = (inputs >= -tail_bound) & (inputs <= tail_bound)
+    outside_interval_mask = ~inside_interval_mask
+
+    outputs = torch.zeros_like(inputs)
+    logabsdet = torch.zeros_like(inputs)
+
+    if tails == 'linear':
+        unnormalized_derivatives = F.pad(unnormalized_derivatives, pad=(1, 1))
+        constant = np.log(np.exp(1 - min_derivative) - 1)
+        unnormalized_derivatives[..., 0] = constant
+        unnormalized_derivatives[..., -1] = constant
+
+        outputs[outside_interval_mask] = inputs[outside_interval_mask]
+        logabsdet[outside_interval_mask] = 0
+    else:
+        raise RuntimeError('{} tails are not implemented.'.format(tails))
+
+    outputs[inside_interval_mask], logabsdet[inside_interval_mask] = rational_quadratic_spline(
+        inputs=inputs[inside_interval_mask],
+        unnormalized_widths=unnormalized_widths[inside_interval_mask, :],
+        unnormalized_heights=unnormalized_heights[inside_interval_mask, :],
+        unnormalized_derivatives=unnormalized_derivatives[inside_interval_mask, :],
+        inverse=inverse,
+        left=-tail_bound, right=tail_bound, bottom=-tail_bound, top=tail_bound,
+        min_bin_width=min_bin_width,
+        min_bin_height=min_bin_height,
+        min_derivative=min_derivative
+    )
+
+    return outputs, logabsdet
+
+def rational_quadratic_spline(inputs,
+                              unnormalized_widths,
+                              unnormalized_heights,
+                              unnormalized_derivatives,
+                              inverse=False,
+                              left=0., right=1., bottom=0., top=1.,
+                              min_bin_width=DEFAULT_MIN_BIN_WIDTH,
+                              min_bin_height=DEFAULT_MIN_BIN_HEIGHT,
+                              min_derivative=DEFAULT_MIN_DERIVATIVE):
+    if torch.min(inputs) < left or torch.max(inputs) > right:
+        raise ValueError('Input to a transform is not within its domain')
+
+    num_bins = unnormalized_widths.shape[-1]
+
+    if min_bin_width * num_bins > 1.0:
+        raise ValueError('Minimal bin width too large for the number of bins')
+    if min_bin_height * num_bins > 1.0:
+        raise ValueError('Minimal bin height too large for the number of bins')
+
+    widths = F.softmax(unnormalized_widths, dim=-1)
+    widths = min_bin_width + (1 - min_bin_width * num_bins) * widths
+    cumwidths = torch.cumsum(widths, dim=-1)
+    cumwidths = F.pad(cumwidths, pad=(1, 0), mode='constant', value=0.0)
+    cumwidths = (right - left) * cumwidths + left
+    cumwidths[..., 0] = left
+    cumwidths[..., -1] = right
+    widths = cumwidths[..., 1:] - cumwidths[..., :-1]
+
+    derivatives = min_derivative + F.softplus(unnormalized_derivatives)
+
+    heights = F.softmax(unnormalized_heights, dim=-1)
+    heights = min_bin_height + (1 - min_bin_height * num_bins) * heights
+    cumheights = torch.cumsum(heights, dim=-1)
+    cumheights = F.pad(cumheights, pad=(1, 0), mode='constant', value=0.0)
+    cumheights = (top - bottom) * cumheights + bottom
+    cumheights[..., 0] = bottom
+    cumheights[..., -1] = top
+    heights = cumheights[..., 1:] - cumheights[..., :-1]
+
+    if inverse:
+        bin_idx = searchsorted(cumheights, inputs)[..., None]
+    else:
+        bin_idx = searchsorted(cumwidths, inputs)[..., None]
+
+    input_cumwidths = cumwidths.gather(-1, bin_idx)[..., 0]
+    input_bin_widths = widths.gather(-1, bin_idx)[..., 0]
+
+    input_cumheights = cumheights.gather(-1, bin_idx)[..., 0]
+    delta = heights / widths
+    input_delta = delta.gather(-1, bin_idx)[..., 0]
+
+    input_derivatives = derivatives.gather(-1, bin_idx)[..., 0]
+    input_derivatives_plus_one = derivatives[..., 1:].gather(-1, bin_idx)[..., 0]
+
+    input_heights = heights.gather(-1, bin_idx)[..., 0]
+
+    if inverse:
+        a = (((inputs - input_cumheights) * (input_derivatives
+                                             + input_derivatives_plus_one
+                                             - 2 * input_delta)
+              + input_heights * (input_delta - input_derivatives)))
+        b = (input_heights * input_derivatives
+             - (inputs - input_cumheights) * (input_derivatives
+                                              + input_derivatives_plus_one
+                                              - 2 * input_delta))
+        c = - input_delta * (inputs - input_cumheights)
+
+        discriminant = b.pow(2) - 4 * a * c
+        assert (discriminant >= 0).all()
+
+        root = (2 * c) / (-b - torch.sqrt(discriminant))
+        outputs = root * input_bin_widths + input_cumwidths
+
+        theta_one_minus_theta = root * (1 - root)
+        denominator = input_delta + ((input_derivatives + input_derivatives_plus_one - 2 * input_delta)
+                                     * theta_one_minus_theta)
+        derivative_numerator = input_delta.pow(2) * (input_derivatives_plus_one * root.pow(2)
+                                                     + 2 * input_delta * theta_one_minus_theta
+                                                     + input_derivatives * (1 - root).pow(2))
+        logabsdet = torch.log(derivative_numerator) - 2 * torch.log(denominator)
+
+        return outputs, -logabsdet
+    else:
+        theta = (inputs - input_cumwidths) / input_bin_widths
+        theta_one_minus_theta = theta * (1 - theta)
+
+        numerator = input_heights * (input_delta * theta.pow(2)
+                                     + input_derivatives * theta_one_minus_theta)
+        denominator = input_delta + ((input_derivatives + input_derivatives_plus_one - 2 * input_delta)
+                                     * theta_one_minus_theta)
+        outputs = input_cumheights + numerator / denominator
+
+        derivative_numerator = input_delta.pow(2) * (input_derivatives_plus_one * theta.pow(2)
+                                                     + 2 * input_delta * theta_one_minus_theta
+                                                     + input_derivatives * (1 - theta).pow(2))
+        logabsdet = torch.log(derivative_numerator) - 2 * torch.log(denominator)
+
+        return outputs, logabsdet

utils.py

@@ -0,0 +1,284 @@
+import os
+import glob
+import sys
+import argparse
+import logging
+import json
+import subprocess
+import numpy as np
+from scipy.io.wavfile import read
+import torch
+
+MATPLOTLIB_FLAG = False
+
+logging.basicConfig(stream=sys.stdout, level=logging.DEBUG)
+logger = logging
+
+
+def load_checkpoint(checkpoint_path, model, optimizer=None, skip_optimizer=False):
+    assert os.path.isfile(checkpoint_path)
+    checkpoint_dict = torch.load(checkpoint_path, map_location='cpu')
+    iteration = checkpoint_dict['iteration']
+    learning_rate = checkpoint_dict['learning_rate']
+    if optimizer is not None and not skip_optimizer and checkpoint_dict['optimizer'] is not None:
+        optimizer.load_state_dict(checkpoint_dict['optimizer'])
+    saved_state_dict = checkpoint_dict['model']
+    if hasattr(model, 'module'):
+        state_dict = model.module.state_dict()
+    else:
+        state_dict = model.state_dict()
+    new_state_dict = {}
+    for k, v in state_dict.items():
+        try:
+            # assert "dec" in k or "disc" in k
+            # print("load", k)
+            new_state_dict[k] = saved_state_dict[k]
+            assert saved_state_dict[k].shape == v.shape, (saved_state_dict[k].shape, v.shape)
+        except:
+            print("error, %s is not in the checkpoint" % k)
+            new_state_dict[k] = v
+    if hasattr(model, 'module'):
+        model.module.load_state_dict(new_state_dict)
+    else:
+        model.load_state_dict(new_state_dict)
+    print("load ")
+    logger.info("Loaded checkpoint '{}' (iteration {})".format(
+        checkpoint_path, iteration))
+    return model, optimizer, learning_rate, iteration
+
+
+def save_checkpoint(model, optimizer, learning_rate, iteration, checkpoint_path):
+    logger.info("Saving model and optimizer state at iteration {} to {}".format(
+        iteration, checkpoint_path))
+    if hasattr(model, 'module'):
+        state_dict = model.module.state_dict()
+    else:
+        state_dict = model.state_dict()
+    torch.save({'model': state_dict,
+                'iteration': iteration,
+                'optimizer': optimizer.state_dict(),
+                'learning_rate': learning_rate}, checkpoint_path)
+
+
+def summarize(writer, global_step, scalars={}, histograms={}, images={}, audios={}, audio_sampling_rate=22050):
+    for k, v in scalars.items():
+        writer.add_scalar(k, v, global_step)
+    for k, v in histograms.items():
+        writer.add_histogram(k, v, global_step)
+    for k, v in images.items():
+        writer.add_image(k, v, global_step, dataformats='HWC')
+    for k, v in audios.items():
+        writer.add_audio(k, v, global_step, audio_sampling_rate)
+
+
+def latest_checkpoint_path(dir_path, regex="G_*.pth"):
+    f_list = glob.glob(os.path.join(dir_path, regex))
+    f_list.sort(key=lambda f: int("".join(filter(str.isdigit, f))))
+    x = f_list[-1]
+    print(x)
+    return x
+
+
+def plot_spectrogram_to_numpy(spectrogram):
+    global MATPLOTLIB_FLAG
+    if not MATPLOTLIB_FLAG:
+        import matplotlib
+        matplotlib.use("Agg")
+        MATPLOTLIB_FLAG = True
+        mpl_logger = logging.getLogger('matplotlib')
+        mpl_logger.setLevel(logging.WARNING)
+    import matplotlib.pylab as plt
+    import numpy as np
+
+    fig, ax = plt.subplots(figsize=(10, 2))
+    im = ax.imshow(spectrogram, aspect="auto", origin="lower",
+                   interpolation='none')
+    plt.colorbar(im, ax=ax)
+    plt.xlabel("Frames")
+    plt.ylabel("Channels")
+    plt.tight_layout()
+
+    fig.canvas.draw()
+    data = np.fromstring(fig.canvas.tostring_rgb(), dtype=np.uint8, sep='')
+    data = data.reshape(fig.canvas.get_width_height()[::-1] + (3,))
+    plt.close()
+    return data
+
+
+def plot_alignment_to_numpy(alignment, info=None):
+    global MATPLOTLIB_FLAG
+    if not MATPLOTLIB_FLAG:
+        import matplotlib
+        matplotlib.use("Agg")
+        MATPLOTLIB_FLAG = True
+        mpl_logger = logging.getLogger('matplotlib')
+        mpl_logger.setLevel(logging.WARNING)
+    import matplotlib.pylab as plt
+    import numpy as np
+
+    fig, ax = plt.subplots(figsize=(6, 4))
+    im = ax.imshow(alignment.transpose(), aspect='auto', origin='lower',
+                   interpolation='none')
+    fig.colorbar(im, ax=ax)
+    xlabel = 'Decoder timestep'
+    if info is not None:
+        xlabel += '\n\n' + info
+    plt.xlabel(xlabel)
+    plt.ylabel('Encoder timestep')
+    plt.tight_layout()
+
+    fig.canvas.draw()
+    data = np.fromstring(fig.canvas.tostring_rgb(), dtype=np.uint8, sep='')
+    data = data.reshape(fig.canvas.get_width_height()[::-1] + (3,))
+    plt.close()
+    return data
+
+
+def load_wav_to_torch(full_path):
+    sampling_rate, data = read(full_path)
+    return torch.FloatTensor(data.astype(np.float32)), sampling_rate
+
+
+def load_filepaths_and_text(filename, split="|"):
+    with open(filename, encoding='utf-8') as f:
+        filepaths_and_text = [line.strip().split(split) for line in f]
+    return filepaths_and_text
+
+
+def get_hparams(init=True):
+    parser = argparse.ArgumentParser()
+    parser.add_argument('-c', '--config', type=str, default="./configs/base.json",
+                        help='JSON file for configuration')
+    parser.add_argument('-m', '--model', type=str, required=True,
+                        help='Model name')
+
+    args = parser.parse_args()
+    model_dir = os.path.join("./logs", args.model)
+
+    if not os.path.exists(model_dir):
+        os.makedirs(model_dir)
+
+    config_path = args.config
+    config_save_path = os.path.join(model_dir, "config.json")
+    if init:
+        with open(config_path, "r") as f:
+            data = f.read()
+        with open(config_save_path, "w") as f:
+            f.write(data)
+    else:
+        with open(config_save_path, "r") as f:
+            data = f.read()
+    config = json.loads(data)
+
+    hparams = HParams(**config)
+    hparams.model_dir = model_dir
+    return hparams
+
+
+def clean_checkpoints(path_to_models='logs/44k/', n_ckpts_to_keep=2, sort_by_time=True):
+    """Freeing up space by deleting saved ckpts
+
+  Arguments:
+  path_to_models    --  Path to the model directory
+  n_ckpts_to_keep   --  Number of ckpts to keep, excluding G_0.pth and D_0.pth
+  sort_by_time      --  True -> chronologically delete ckpts
+                        False -> lexicographically delete ckpts
+  """
+    import re
+    ckpts_files = [f for f in os.listdir(path_to_models) if os.path.isfile(os.path.join(path_to_models, f))]
+    name_key = (lambda _f: int(re.compile('._(\d+)\.pth').match(_f).group(1)))
+    time_key = (lambda _f: os.path.getmtime(os.path.join(path_to_models, _f)))
+    sort_key = time_key if sort_by_time else name_key
+    x_sorted = lambda _x: sorted([f for f in ckpts_files if f.startswith(_x) and not f.endswith('_0.pth')],
+                                 key=sort_key)
+    to_del = [os.path.join(path_to_models, fn) for fn in
+              (x_sorted('G')[:-n_ckpts_to_keep] + x_sorted('D')[:-n_ckpts_to_keep])]
+    del_info = lambda fn: logger.info(f".. Free up space by deleting ckpt {fn}")
+    del_routine = lambda x: [os.remove(x), del_info(x)]
+    rs = [del_routine(fn) for fn in to_del]
+
+def get_hparams_from_dir(model_dir):
+    config_save_path = os.path.join(model_dir, "config.json")
+    with open(config_save_path, "r") as f:
+        data = f.read()
+    config = json.loads(data)
+
+    hparams = HParams(**config)
+    hparams.model_dir = model_dir
+    return hparams
+
+
+def get_hparams_from_file(config_path):
+    with open(config_path, "r") as f:
+        data = f.read()
+    config = json.loads(data)
+
+    hparams = HParams(**config)
+    return hparams
+
+
+def check_git_hash(model_dir):
+    source_dir = os.path.dirname(os.path.realpath(__file__))
+    if not os.path.exists(os.path.join(source_dir, ".git")):
+        logger.warn("{} is not a git repository, therefore hash value comparison will be ignored.".format(
+            source_dir
+        ))
+        return
+
+    cur_hash = subprocess.getoutput("git rev-parse HEAD")
+
+    path = os.path.join(model_dir, "githash")
+    if os.path.exists(path):
+        saved_hash = open(path).read()
+        if saved_hash != cur_hash:
+            logger.warn("git hash values are different. {}(saved) != {}(current)".format(
+                saved_hash[:8], cur_hash[:8]))
+    else:
+        open(path, "w").write(cur_hash)
+
+
+def get_logger(model_dir, filename="train.log"):
+    global logger
+    logger = logging.getLogger(os.path.basename(model_dir))
+    logger.setLevel(logging.DEBUG)
+
+    formatter = logging.Formatter("%(asctime)s\t%(name)s\t%(levelname)s\t%(message)s")
+    if not os.path.exists(model_dir):
+        os.makedirs(model_dir)
+    h = logging.FileHandler(os.path.join(model_dir, filename))
+    h.setLevel(logging.DEBUG)
+    h.setFormatter(formatter)
+    logger.addHandler(h)
+    return logger
+
+
+class HParams():
+    def __init__(self, **kwargs):
+        for k, v in kwargs.items():
+            if type(v) == dict:
+                v = HParams(**v)
+            self[k] = v
+
+    def keys(self):
+        return self.__dict__.keys()
+
+    def items(self):
+        return self.__dict__.items()
+
+    def values(self):
+        return self.__dict__.values()
+
+    def __len__(self):
+        return len(self.__dict__)
+
+    def __getitem__(self, key):
+        return getattr(self, key)
+
+    def __setitem__(self, key, value):
+        return setattr(self, key, value)
+
+    def __contains__(self, key):
+        return key in self.__dict__
+
+    def __repr__(self):
+        return self.__dict__.__repr__()