text cleaner from https://github.com/CjangCjengh/vits

original repo: https://github.com/jaywalnut310/vits

See vits-finetuning

(Suggestion) Python == 3.7

Only Japanese datasets can be used for fine-tuning in this repo.

git clone https://github.com/SayaSS/vits-finetuning.git

pip install -r requirements.txt

• G_0.pth
• D_0.pth
• Edit "model_dir"(line 152) in utils.py
• Put pre-trained models in the "model_dir"/checkpoints

• G_0-p.pth
• D_0-p.pth

• Speaker ID should be between 0-803.
• About 50 audio-text pairs will suffice and 100-600 epochs could have quite good performance, but more data may be better.
• Resample all audio to 22050Hz, 16-bit, mono wav files.
• Audio files should be >=1s and <=10s.

path/to/XXX.wav|speaker id|transcript

• Example

dataset/001.wav|10|こんにちは。

For complete examples, please see filelists/miyu_train.txt and filelists/miyu_val.txt.

python preprocess.py --filelists path/to/filelist_train.txt path/to/filelist_val.txt

Edit "training_files" and "validation_files" in configs/config.json

cd monotonic_align
python setup.py build_ext --inplace
cd ..

# Mutiple speakers
python train_ms.py -c configs/config.json -m checkpoints