tuna2134/sbv2-v2
1
0
Fork 0
Code Issues Pull Requests 1 Actions Packages Projects Releases Wiki Activity

Remove: modules under common/ that have been rewritten

Browse Source
This commit is contained in:
tsukumi
2024-03-06 22:29:12 +00:00
parent 9546495434
commit f880641eb5
27 changed files with 44 additions and 129 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
app.py
View File

@@ -10,7 +10,7 @@ import gradio as gr
import torch
import yaml
from common.constants import (
from style_bert_vits2.constants import (
DEFAULT_ASSIST_TEXT_WEIGHT,
DEFAULT_LENGTH,
DEFAULT_LINE_SPLIT,
@@ -21,10 +21,10 @@ from common.constants import (
DEFAULT_STYLE,
DEFAULT_STYLE_WEIGHT,
GRADIO_THEME,
LATEST_VERSION,
VERSION,
Languages,
)
from common.log import logger
from style_bert_vits2.logging import logger
from common.tts_model import ModelHolder
from infer import InvalidToneError
from text.japanese import g2kata_tone, kata_tone2phone_tone, text_normalize
@@ -202,7 +202,7 @@ examples = [
]
initial_md = f"""
# Style-Bert-VITS2 ver {LATEST_VERSION} 音声合成
# Style-Bert-VITS2 ver {VERSION} 音声合成
- Ver 2.3で追加されたエディターのほうが実際に読み上げさせるには使いやすいかもしれません。`Editor.bat`か`python server_editor.py`で起動できます。

2
attentions.py
View File

@@ -4,7 +4,7 @@ from torch import nn
from torch.nn import functional as F
from style_bert_vits2.models import commons
from common.log import logger as logging
from style_bert_vits2.logging import logger as logging
class LayerNorm(nn.Module):

4
bert_gen.py
View File

@@ -7,8 +7,8 @@ from tqdm import tqdm
from style_bert_vits2.models import commons
import utils
from common.log import logger
from common.stdout_wrapper import SAFE_STDOUT
from style_bert_vits2.logging import logger
from style_bert_vits2.utils.stdout_wrapper import SAFE_STDOUT
from config import config
from text import cleaned_text_to_sequence, get_bert

28
common/constants.py
View File

@@ -1,28 +0,0 @@
import enum
# Built-in theme: "default", "base", "monochrome", "soft", "glass"
# See https://huggingface.co/spaces/gradio/theme-gallery for more themes
GRADIO_THEME: str = "NoCrypt/miku"
LATEST_VERSION: str = "2.3.1"
USER_DICT_DIR = "dict_data"
DEFAULT_STYLE: str = "Neutral"
DEFAULT_STYLE_WEIGHT: float = 5.0
class Languages(str, enum.Enum):
JP = "JP"
EN = "EN"
ZH = "ZH"
DEFAULT_SDP_RATIO: float = 0.2
DEFAULT_NOISE: float = 0.6
DEFAULT_NOISEW: float = 0.8
DEFAULT_LENGTH: float = 1.0
DEFAULT_LINE_SPLIT: bool = True
DEFAULT_SPLIT_INTERVAL: float = 0.5
DEFAULT_ASSIST_TEXT_WEIGHT: float = 0.7
DEFAULT_ASSIST_TEXT_WEIGHT: float = 1.0

17
common/log.py
View File

@@ -1,17 +0,0 @@
"""
logger封装
"""
from loguru import logger
from .stdout_wrapper import SAFE_STDOUT
# 移除所有默认的处理器
logger.remove()
# 自定义格式并添加到标准输出
log_format = (
"<g>{time:MM-DD HH:mm:ss}</g> |<lvl>{level:^8}</lvl>| {file}:{line} | {message}"
)
logger.add(SAFE_STDOUT, format=log_format, backtrace=True, diagnose=True)

40
common/stdout_wrapper.py
View File

@@ -1,40 +0,0 @@
"""
`sys.stdout` wrapper for both Google Colab and local environment.
"""
import sys
import tempfile
class StdoutWrapper:
def __init__(self):
self.temp_file = tempfile.NamedTemporaryFile(
mode="w+", delete=False, encoding="utf-8"
)
self.original_stdout = sys.stdout
def write(self, message: str):
self.temp_file.write(message)
self.temp_file.flush()
print(message, end="", file=self.original_stdout)
def flush(self):
self.temp_file.flush()
def read(self):
self.temp_file.seek(0)
return self.temp_file.read()
def close(self):
self.temp_file.close()
def fileno(self):
return self.temp_file.fileno()
try:
import google.colab
SAFE_STDOUT = StdoutWrapper()
except ImportError:
SAFE_STDOUT = sys.stdout

2
config.py
View File

@@ -9,7 +9,7 @@ from typing import Dict, List
import torch
import yaml
from common.log import logger
from style_bert_vits2.logging import logger
# If not cuda available, set possible devices to cpu
cuda_available = torch.cuda.is_available()

2
data_utils.py
View File

@@ -11,7 +11,7 @@ from style_bert_vits2.models import commons
from config import config
from mel_processing import mel_spectrogram_torch, spectrogram_torch
from text import cleaned_text_to_sequence
from common.log import logger
from style_bert_vits2.logging import logger
from utils import load_filepaths_and_text, load_wav_to_torch
"""Multi speaker version"""

4
default_style.py
View File

@@ -1,6 +1,6 @@
import os
from common.log import logger
from common.constants import DEFAULT_STYLE
from style_bert_vits2.logging import logger
from style_bert_vits2.constants import DEFAULT_STYLE
import numpy as np
import json

2
infer.py
View File

@@ -7,7 +7,7 @@ from models_jp_extra import SynthesizerTrn as SynthesizerTrnJPExtra
from text import cleaned_text_to_sequence, get_bert
from text.cleaner import clean_text
from text.symbols import symbols
from common.log import logger
from style_bert_vits2.logging import logger
class InvalidToneError(ValueError):

2
initialize.py
View File

@@ -5,7 +5,7 @@ from pathlib import Path
import yaml
from huggingface_hub import hf_hub_download
from common.log import logger
from style_bert_vits2.logging import logger
def download_bert_models():

2
losses.py
View File

@@ -2,7 +2,7 @@ import torch
import torchaudio
from transformers import AutoModel
from common.log import logger
from style_bert_vits2.logging import logger
def feature_loss(fmap_r, fmap_g):

4
preprocess_text.py
View File

@@ -7,8 +7,8 @@ from typing import Optional
import click
from tqdm import tqdm
from common.log import logger
from common.stdout_wrapper import SAFE_STDOUT
from style_bert_vits2.logging import logger
from style_bert_vits2.utils.stdout_wrapper import SAFE_STDOUT
from config import config
from text.cleaner import clean_text

4
resample.py
View File

@@ -7,8 +7,8 @@ import pyloudnorm as pyln
import soundfile
from tqdm import tqdm
from common.log import logger
from common.stdout_wrapper import SAFE_STDOUT
from style_bert_vits2.logging import logger
from style_bert_vits2.utils.stdout_wrapper import SAFE_STDOUT
from config import config
DEFAULT_BLOCK_SIZE: float = 0.400 # seconds

4
server_fastapi.py
View File

@@ -20,7 +20,7 @@ from fastapi.middleware.cors import CORSMiddleware
from fastapi.responses import FileResponse, Response
from scipy.io import wavfile
from common.constants import (
from style_bert_vits2.constants import (
DEFAULT_ASSIST_TEXT_WEIGHT,
DEFAULT_LENGTH,
DEFAULT_LINE_SPLIT,
@@ -32,7 +32,7 @@ from common.constants import (
DEFAULT_STYLE_WEIGHT,
Languages,
)
from common.log import logger
from style_bert_vits2.logging import logger
from common.tts_model import Model, ModelHolder
from config import config

4
slice.py
View File

@@ -8,8 +8,8 @@ import torch
import yaml
from tqdm import tqdm
from common.log import logger
from common.stdout_wrapper import SAFE_STDOUT
from style_bert_vits2.logging import logger
from style_bert_vits2.utils.stdout_wrapper import SAFE_STDOUT
vad_model, utils = torch.hub.load(
repo_or_dir="snakers4/silero-vad",

2
speech_mos.py
View File

@@ -10,7 +10,7 @@ import pandas as pd
import torch
from tqdm import tqdm
from common.log import logger
from style_bert_vits2.logging import logger
from common.tts_model import Model
from config import config

4
style_gen.py
View File

@@ -7,8 +7,8 @@ import torch
from tqdm import tqdm
import utils
from common.log import logger
from common.stdout_wrapper import SAFE_STDOUT
from style_bert_vits2.logging import logger
from style_bert_vits2.utils.stdout_wrapper import SAFE_STDOUT
from config import config
warnings.filterwarnings("ignore", category=UserWarning)

2
text/japanese.py
View File

@@ -8,7 +8,7 @@ import pyopenjtalk
from num2words import num2words
from transformers import AutoTokenizer
from common.log import logger
from style_bert_vits2.logging import logger
from text import punctuation
from text.japanese_mora_list import (
mora_kata_to_mora_phonemes,

4
train_ms.py
View File

@@ -18,8 +18,8 @@ from tqdm import tqdm
from style_bert_vits2.models import commons
import default_style
import utils
from common.log import logger
from common.stdout_wrapper import SAFE_STDOUT
from style_bert_vits2.logging import logger
from style_bert_vits2.utils.stdout_wrapper import SAFE_STDOUT
from config import config
from data_utils import (
DistributedBucketSampler,

4
train_ms_jp_extra.py
View File

@@ -18,8 +18,8 @@ from huggingface_hub import HfApi
from style_bert_vits2.models import commons
import default_style
import utils
from common.log import logger
from common.stdout_wrapper import SAFE_STDOUT
from style_bert_vits2.logging import logger
from style_bert_vits2.utils.stdout_wrapper import SAFE_STDOUT
from config import config
from data_utils import (
DistributedBucketSampler,

6
transcribe.py
View File

@@ -7,9 +7,9 @@ import yaml
from faster_whisper import WhisperModel
from tqdm import tqdm
from common.constants import Languages
from common.log import logger
from common.stdout_wrapper import SAFE_STDOUT
from style_bert_vits2.constants import Languages
from style_bert_vits2.logging import logger
from style_bert_vits2.utils.stdout_wrapper import SAFE_STDOUT
def transcribe(wav_path: Path, initial_prompt=None, language="ja"):

2
utils.py
View File

@@ -13,7 +13,7 @@ from safetensors import safe_open
from safetensors.torch import save_file
from scipy.io.wavfile import read
from common.log import logger
from style_bert_vits2.logging import logger
MATPLOTLIB_FLAG = False

4
webui_dataset.py
View File

@@ -4,8 +4,8 @@ import os
import gradio as gr
import yaml
from common.constants import GRADIO_THEME
from common.log import logger
from style_bert_vits2.constants import GRADIO_THEME
from style_bert_vits2.logging import logger
from common.subprocess_utils import run_script_with_log
# Get path settings

4
webui_merge.py
View File

@@ -11,8 +11,8 @@ import yaml
from safetensors import safe_open
from safetensors.torch import save_file
from common.constants import DEFAULT_STYLE, GRADIO_THEME
from common.log import logger
from style_bert_vits2.constants import DEFAULT_STYLE, GRADIO_THEME
from style_bert_vits2.logging import logger
from common.tts_model import Model, ModelHolder
voice_keys = ["dec"]

4
webui_style_vectors.py
View File

@@ -12,8 +12,8 @@ from sklearn.cluster import DBSCAN, AgglomerativeClustering, KMeans
from sklearn.manifold import TSNE
from umap import UMAP
from common.constants import DEFAULT_STYLE, GRADIO_THEME
from common.log import logger
from style_bert_vits2.constants import DEFAULT_STYLE, GRADIO_THEME
from style_bert_vits2.logging import logger
from config import config
# Get path settings

8
webui_train.py
View File

@@ -14,9 +14,9 @@ from pathlib import Path
import gradio as gr
import yaml
from common.constants import GRADIO_THEME, LATEST_VERSION
from common.log import logger
from common.stdout_wrapper import SAFE_STDOUT
from style_bert_vits2.constants import GRADIO_THEME, VERSION
from style_bert_vits2.logging import logger
from style_bert_vits2.utils.stdout_wrapper import SAFE_STDOUT
from common.subprocess_utils import run_script_with_log, second_elem_of
logger_handler = None
@@ -399,7 +399,7 @@ def run_tensorboard(model_name):
initial_md = f"""
# Style-Bert-VITS2 ver {LATEST_VERSION} 学習用WebUI
# Style-Bert-VITS2 ver {VERSION} 学習用WebUI
## 使い方