Merge branch 'dev' of github.com:litagin02/Style-Bert-VITS2

gradio_tabs/dataset.py

@@ -177,9 +177,16 @@ def create_dataset_app() -> gr.Blocks:
                     value=False,
                 )
                 hf_repo_id = gr.Dropdown(
-                    ["openai/whisper", "kotoba-tech/kotoba-whisper-v1.1"],
+                    [
-                    label="HuggingFaceのWhisperモデル",
+                        "openai/whisper-large-v3-turbo",
-                    value="openai/whisper",
+                        "openai/whisper-large-v3",
+                        "openai/whisper-large-v2",
+                        "kotoba-tech/kotoba-whisper-v1.1",
+                        "kotoba-tech/kotoba-whisper-v2.1",
+                        "litagin/anime-whisper",
+                    ],
+                    label="HuggingFaceのWhisper repo_id",
+                    value="openai/whisper-large-v3",
                     visible=False,
                 )
                 compute_type = gr.Dropdown(

gradio_tabs/merge.py

@@ -897,7 +897,7 @@ initial_md = """
 - 話者数が違うモデル同士はおそらくマージできません。
 """
 
-style_merge_md = f"""
+style_merge_md = """
 ## 3. スタイルベクトルのマージ
 
 1. マージ後のモデルにいくつスタイルを追加したいかを「作りたいスタイル数」で指定

preprocess_text.py

@@ -2,7 +2,7 @@ import argparse
 import json
 from collections import defaultdict
 from pathlib import Path
-from random import sample, shuffle
+from random import sample
 from typing import Optional
 
 from tqdm import tqdm

style_bert_vits2/models/infer_onnx.py

@@ -1,4 +1,5 @@
-from typing import Any, Optional, Sequence, Union
+from collections.abc import Sequence
+from typing import Any, Optional, Union
 
 import numpy as np
 import onnxruntime

style_bert_vits2/models/utils/__init__.py

@@ -11,8 +11,10 @@ import torch
 from numpy.typing import NDArray
 
 from style_bert_vits2.logging import logger
-from style_bert_vits2.models.utils import checkpoints  # type: ignore
+from style_bert_vits2.models.utils import (
-from style_bert_vits2.models.utils import safetensors  # type: ignore
+    checkpoints,  # type: ignore # noqa: F401
+    safetensors,  # type: ignore # noqa: F401
+)
 
 
 if TYPE_CHECKING:

style_bert_vits2/nlp/__init__.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
-from typing import TYPE_CHECKING, Any, Optional, Sequence, Union
+from collections.abc import Sequence
+from typing import TYPE_CHECKING, Any, Optional, Union
 
 from numpy.typing import NDArray
 

style_bert_vits2/nlp/chinese/bert_feature.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
-from typing import TYPE_CHECKING, Any, Optional, Sequence, Union
+from collections.abc import Sequence
+from typing import TYPE_CHECKING, Any, Optional, Union
 
 import numpy as np
 import onnxruntime

style_bert_vits2/nlp/english/bert_feature.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
-from typing import TYPE_CHECKING, Any, Optional, Sequence, Union
+from collections.abc import Sequence
+from typing import TYPE_CHECKING, Any, Optional, Union
 
 import numpy as np
 import onnxruntime

style_bert_vits2/nlp/japanese/bert_feature.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
-from typing import TYPE_CHECKING, Any, Optional, Sequence, Union
+from collections.abc import Sequence
+from typing import TYPE_CHECKING, Any, Optional, Union
 
 import numpy as np
 import onnxruntime

style_bert_vits2/nlp/onnx_bert_models.py

@@ -11,8 +11,9 @@ Style-Bert-VITS2 の ONNX 推論に必要な各言語ごとの ONNX 版 BERT モ
 
 import gc
 import time
+from collections.abc import Sequence
 from pathlib import Path
-from typing import Any, Optional, Sequence, Union
+from typing import Any, Optional, Union
 
 import onnxruntime
 from huggingface_hub import hf_hub_download

style_bert_vits2/tts_model.py

@@ -2,8 +2,9 @@ from __future__ import annotations
 
 import gc
 import time
+from collections.abc import Sequence
 from pathlib import Path
-from typing import TYPE_CHECKING, Any, Optional, Sequence, Union
+from typing import TYPE_CHECKING, Any, Optional, Union
 
 import numpy as np
 import onnxruntime

style_bert_vits2/utils/__init__.py

@@ -1,4 +1,5 @@
-from typing import Any, Sequence, Union
+from collections.abc import Sequence
+from typing import Any, Union
 
 import onnxruntime
 

tests/test_main.py

@@ -1,4 +1,5 @@
-from typing import Any, Literal, Sequence
+from collections.abc import Sequence
+from typing import Any, Literal
 
 import pytest
 from scipy.io import wavfile

transcribe.py

@@ -67,8 +67,7 @@ def transcribe_files_with_hf_whisper(
         "num_beams": num_beams,
         "no_repeat_ngram_size": no_repeat_ngram_size,
     }
-    logger.info(f"generate_kwargs: {generate_kwargs}")
+    logger.info(f"generate_kwargs: {generate_kwargs}, loading pipeline...")
-
     pipe = pipeline(
         model=model_id,
         max_new_tokens=128,
@@ -79,6 +78,7 @@ def transcribe_files_with_hf_whisper(
         trust_remote_code=True,
         # generate_kwargs=generate_kwargs,
     )
+    logger.info("Loaded pipeline")
     if initial_prompt is not None:
         prompt_ids: torch.Tensor = pipe.tokenizer.get_prompt_ids(
             initial_prompt, return_tensors="pt"
@@ -196,10 +196,7 @@ if __name__ == "__main__":
             with open(output_file, "a", encoding="utf-8") as f:
                 f.write(f"{wav_rel_path}|{model_name}|{language_id}|{text}\n")
     else:
-        if args.hf_repo_id == "":
+        model_id = args.hf_repo_id
-            model_id = f"openai/whisper-{args.model}"
-        else:
-            model_id = args.hf_repo_id
         logger.info(f"Loading HF Whisper model ({model_id})")
         pbar = tqdm(total=len(wav_files), file=SAFE_STDOUT)
         results = transcribe_files_with_hf_whisper(

vad_filter.py

@@ -1,92 +0,0 @@
-import argparse
-import os
-import shutil
-import sys
-from pathlib import Path
-
-import pandas as pd
-import torch
-from tqdm import tqdm
-
-from style_bert_vits2.logging import logger
-
-
-vad_model, utils = torch.hub.load(
-    repo_or_dir="litagin02/silero-vad",
-    model="silero_vad",
-    onnx=True,
-    trust_repo=True,
-)
-
-(get_speech_timestamps, _, read_audio, *_) = utils
-
-
-def get_speech_ratio(audio_file):
-    sampling_rate = 16000
-
-    wav = read_audio(audio_file, sampling_rate=sampling_rate)
-    speech_timestamps = get_speech_timestamps(
-        wav, vad_model, sampling_rate=sampling_rate
-    )
-
-    speech_dur_ms = 0
-
-    for ts in speech_timestamps:
-        start_ms = ts["start"] / 16
-        end_ms = ts["end"] / 16
-        speech_dur_ms += end_ms - start_ms
-
-    total_dur_ms = len(wav) / sampling_rate * 1000
-    return speech_dur_ms / total_dur_ms
-
-
-def process(file: Path):
-    speech_ratio = get_speech_ratio(file)
-    return file, speech_ratio
-
-
-def main():
-    parser = argparse.ArgumentParser(description="Calculate speech ratio.")
-    parser.add_argument(
-        "-i", "--input", help="Directory containing audio files", required=True
-    )
-    args = parser.parse_args()
-
-    if os.path.exists(os.path.join(args.input, "low_speech_ratio")):
-        logger.info("Low speech ratio directory already exists, skipping...")
-        exit(0)
-
-    data_dir = Path(args.input)
-    wav_files = list(data_dir.glob("*.wav"))
-    wav_files.sort()
-
-    if len(wav_files) < 100:
-        logger.warning("Too few files, skipping...")
-        exit(0)
-
-    logger.info(f"Start VAD filtering for {data_dir}...")
-
-    results = []
-
-    for wav_file in tqdm(wav_files, file=sys.stdout):
-        speech_ratio = get_speech_ratio(wav_file)
-        results.append((wav_file, speech_ratio))
-
-    results_df = pd.DataFrame(results, columns=["file", "speech_ratio"])
-    results_df.to_csv(os.path.join(data_dir, "speech_ratio.csv"), index=False)
-
-    logger.info(f"Speech ratio stats:\n{results_df['speech_ratio'].describe()}")
-    threshold = 0.5
-
-    low_speech_ratio_dir = os.path.join(data_dir, "low_speech_ratio")
-    os.makedirs(low_speech_ratio_dir, exist_ok=True)
-
-    low_speech_files = results_df[results_df["speech_ratio"] < threshold]["file"]
-    logger.info(f"Moving {len(low_speech_files)} files to {low_speech_ratio_dir}...")
-    for low_speech_file in low_speech_files:
-        shutil.move(low_speech_file, low_speech_ratio_dir)
-    logger.success("VAD filtering completed.")
-
-
-if __name__ == "__main__":
-    main()