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

Add jvnv jp models, ver 2.0.1

Browse Source
This commit is contained in:
litagin02
2024-02-05 22:08:18 +09:00
parent 204aed69ad
commit 1cf7599f80
8 changed files with 92 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
README.md
View File

@@ -1,11 +1,12 @@
# 2024-02-03: JP-Extra対応版 ver 2.0をリリースしました！
[**ver 2.0 リリースページ**](https://github.com/litagin02/Style-Bert-VITS2/releases/tag/2.0)
[**リリースページ**](https://github.com/litagin02/Style-Bert-VITS2/releases/tag/2.0)
モデルとして [Bert-VITS2 JP-Extra](https://github.com/fishaudio/Bert-VITS2/releases/tag/JP-Exta) を元としたものが使えるようになりました。日本語の発音やアクセントや自然性が大幅に向上する傾向があります。
[**Changelog**](docs/CHANGELOG.md)
- 2024-02-05: ver 2.0.1
- 2024-02-03: ver 2.0
- 2024-01-09: ver 1.3
- 2023-12-31: ver 1.2

2
configs/config.json
View File

@@ -67,5 +67,5 @@
"use_spectral_norm": false,
"gin_channels": 256
},
"version": "2.0"
"version": "2.0.1"
}

2
configs/configs_jp_extra.json
View File

@@ -74,5 +74,5 @@
"initial_channel": 64
}
},
"version": "2.0-JP-Extra"
"version": "2.0.1-JP-Extra"
}

8
docs/CHANGELOG.md
View File

@@ -1,5 +1,13 @@
# Changelog
## v2.0.1 (2024-02-05)
軽微なバグ修正や改善
- スタイルベクトルに`nan`が含まれていた場合（主に音声ファイルが極端に短い場合に発生）、それを学習リストから除外するように修正
- colabにマージの追加
- 学習時のプログレスバーの表示がおかしかったのを修正
- デフォルトのjvnvモデルをJP-Extra版にアップデート。
## v2.0 (2024-02-03)
### 大きい変更

2
docs/README_en.md
View File

@@ -1,3 +1,5 @@
# This English README is for 1.x versions. WIP for 2.x versions.
# Style-Bert-VITS2
Bert-VITS2 with more controllable voice styles.

24
initialize.py
View File

@@ -66,18 +66,18 @@ def download_jp_extra_pretrained_models():
def download_jvnv_models():
files = [
"jvnv-F1/config.json",
"jvnv-F1/jvnv-F1.safetensors",
"jvnv-F1/style_vectors.npy",
"jvnv-F2/config.json",
"jvnv-F2/jvnv-F2.safetensors",
"jvnv-F2/style_vectors.npy",
"jvnv-M1/config.json",
"jvnv-M1/jvnv-M1.safetensors",
"jvnv-M1/style_vectors.npy",
"jvnv-M2/config.json",
"jvnv-M2/jvnv-M2.safetensors",
"jvnv-M2/style_vectors.npy",
"jvnv--jp/config.json",
"jvnv-F1-jp/jvnv-F1-jp_e160_s14000.safetensors",
"jvnv-F1-jp/style_vectors.npy",
"jvnv-F2-jp/config.json",
"jvnv-F2-jp/jvnv-F2_e166_s20000.safetensors",
"jvnv-F2-jp/style_vectors.npy",
"jvnv-M1-jp/config.json",
"jvnv-M1-jp/jvnv-M1-jp_e158_s14000.safetensors",
"jvnv-M1-jp/style_vectors.npy",
"jvnv-M2-jp/config.json",
"jvnv-M2-jp/jvnv-M2-jp_e159_s17000.safetensors",
"jvnv-M2-jp/style_vectors.npy",
]
for file in files:
if not Path(f"model_assets/{file}").exists():

2
scripts/Update-to-JP-Extra.bat
View File

@@ -42,7 +42,7 @@ pushd Style-Bert-VITS2
@REM 初期化（必要なモデルのダウンロード）
python initialize.py
echo Style-Bert-VITS2の2.0へのアップデートが完了しました。
echo Style-Bert-VITS2の2.0.1へのアップデートが完了しました。
pause

81
style_gen.py
View File

@@ -1,5 +1,5 @@
import argparse
import concurrent.futures
from concurrent.futures import ThreadPoolExecutor
import warnings
import numpy as np
@@ -20,6 +20,12 @@ device = torch.device(config.style_gen_config.device)
inference.to(device)
class NaNValueError(ValueError):
"""カスタム例外クラス。NaN値が見つかった場合に使用されます。"""
pass
# 推論時にインポートするために短いが関数を書く
def get_style_vector(wav_path):
return inference(wav_path)
@@ -29,10 +35,22 @@ def save_style_vector(wav_path):
try:
style_vec = get_style_vector(wav_path)
except Exception as e:
logger.error(f"\nError occurred with file: {wav_path}, Details:\n{e}\n")
logger.error(f"Error occurred with file: {wav_path}, Details:\n{e}\n")
raise
# 値にNaNが含まれていると悪影響なのでチェックする
if np.isnan(style_vec).any():
logger.warning(f"NaN value found in style vector: {wav_path}")
raise NaNValueError(f"NaN value found in style vector: {wav_path}")
np.save(f"{wav_path}.npy", style_vec) # `test.wav` -> `test.wav.npy`
return style_vec
def process_line(line):
wavname = line.split("|")[0]
try:
save_style_vector(wavname)
return line, None
except NaNValueError:
return line, "nan_error"
def save_average_style_vector(style_vectors, filename="style_vectors.npy"):
@@ -56,22 +74,53 @@ if __name__ == "__main__":
device = config.style_gen_config.device
lines = []
training_lines = []
with open(hps.data.training_files, encoding="utf-8") as f:
lines.extend(f.readlines())
with open(hps.data.validation_files, encoding="utf-8") as f:
lines.extend(f.readlines())
wavnames = [line.split("|")[0] for line in lines]
with concurrent.futures.ThreadPoolExecutor(max_workers=num_processes) as executor:
list(
training_lines.extend(f.readlines())
with ThreadPoolExecutor(max_workers=num_processes) as executor:
training_results = list(
tqdm(
executor.map(save_style_vector, wavnames),
total=len(wavnames),
executor.map(process_line, training_lines),
total=len(training_lines),
file=SAFE_STDOUT,
)
)
ok_training_lines = [line for line, error in training_results if error is None]
nan_training_lines = [
line for line, error in training_results if error == "nan_error"
]
if nan_training_lines:
nan_files = [line.split("|")[0] for line in nan_training_lines]
logger.warning(
f"Found NaN value in {len(nan_training_lines)} files: {nan_files}, so they will be deleted from training data."
)
logger.info(f"Finished generating style vectors! total: {len(wavnames)} npy files.")
val_lines = []
with open(hps.data.validation_files, encoding="utf-8") as f:
val_lines.extend(f.readlines())
with ThreadPoolExecutor(max_workers=num_processes) as executor:
val_results = list(
tqdm(
executor.map(process_line, val_lines),
total=len(val_lines),
file=SAFE_STDOUT,
)
)
ok_val_lines = [line for line, error in val_results if error is None]
nan_val_lines = [line for line, error in val_results if error == "nan_error"]
if nan_val_lines:
nan_files = [line.split("|")[0] for line in nan_val_lines]
logger.warning(
f"Found NaN value in {len(nan_val_lines)} files: {nan_files}, so they will be deleted from validation data."
)
with open(hps.data.training_files, "w", encoding="utf-8") as f:
f.writelines(ok_training_lines)
with open(hps.data.validation_files, "w", encoding="utf-8") as f:
f.writelines(ok_val_lines)
ok_num = len(ok_training_lines) + len(ok_val_lines)
logger.info(f"Finished generating style vectors! total: {ok_num} npy files.")