/VITS-fast-fine-tuning

This repo is a pipeline of VITS finetuning for fast speaker adaptation TTS, and any-to-any voice conversion

Primary LanguagePythonApache License 2.0Apache-2.0

中文文档请点击这里

VITS Fast Fine-tuning

This repo will guide you to add your own character voices, or even your own voice, into existing VITS TTS model to make it able to do the following tasks in less than 1 hour:

  1. Many-to-many voice conversion between any characters you added & preset characters in the model.
  2. English, Japanese & Chinese Text-to-Speech synthesis with the characters you added & preset characters

Welcome to play around with the base models!
Chinese & English & Japanese：Hugging Face Spaces Author: Me

Chinese & Japanese：Hugging Face Spaces Author: SayaSS

Currently Supported Tasks:

  • Clone character voice from 10+ short audios
  • Clone character voice from long audio(s) >= 3 minutes (one audio should contain single speaker only)
  • Clone character voice from videos(s) >= 3 minutes (one video should contain single speaker only)
  • Clone character voice from BILIBILI video links (one video should contain single speaker only)

Currently Supported Characters for TTS & VC:

  • Any character you wish as long as you have their voices! (Note that voice conversion can only be conducted between any two speakers in the model)

Fine-tuning

It's recommended to perform fine-tuning on Google Colab because the original VITS has some dependencies that are difficult to configure.

How long does it take?

  1. Install dependencies (3 min)
  2. Choose pretrained model to start. The detailed differences between them are described in Colab Notebook
  3. Upload the voice samples of the characters you wish to add，see DATA.MD for detailed uploading options.
  4. Start fine-tuning. Time taken varies from 20 minutes ~ 2 hours, depending on the number of voices you uploaded.

Inference or Usage (Currently support Windows only)

  1. Remember to download your fine-tuned model!
  2. Download the latest release
  3. Put your model & config file into the folder inference, which are named G_latest.pth and finetune_speaker.json, respectively.
  4. The file structure should be as follows:
inference
├───inference.exe
├───...
├───finetune_speaker.json
└───G_latest.pth
  1. run inference.exe, the browser should pop up automatically.

Use in MoeGoe

  1. Prepare downloaded model & config file, which are named G_latest.pth and moegoe_config.json, respectively.
  2. Follow MoeGoe page instructions to install, configure path, and use.