2024-01-16 18:02:51 +08:00
2024-01-16 17:39:23 +08:00
2024-01-16 17:25:53 +08:00
2024-01-16 17:35:12 +08:00
2024-01-16 17:25:53 +08:00
2024-01-16 17:25:53 +08:00
2024-01-15 02:05:22 +08:00
2024-01-16 18:02:51 +08:00
2024-01-16 17:36:27 +08:00

I am organizing and uploading the codes. It will be public in one day.

requirment (How to install)

python and pytorch version

py39+pytorch2.0.1+cu11 passed the test.

pip packages

pip install torch numpy scipy tensorboard librosa==0.9.2 numba==0.56.4 pytorch-lightning gradio==3.14.0 ffmpeg-python onnxruntime tqdm==4.59.0 cn2an pypinyin pyopenjtalk g2p_en

additionally

If you need the Chinese ASR feature supported by funasr, you should

pip install modelscope sentencepiece funasr

You need ffmpeg.

Ubuntu/Debian users

sudo apt install ffmpeg

MacOS users

brew install ffmpeg

Windows users

download and put it to the GPT-SoVITS root.

Description
1 min voice data can also be used to train a good TTS model! (few shot voice cloning)
Readme MIT 28 MiB
Languages
Python 97.5%
Jupyter Notebook 1%
Cuda 0.6%
C 0.4%
Shell 0.3%