Jason-Lu's picture
Update README.md
22ca975
|
raw
history blame
771 Bytes
metadata
license: cc-by-nc-4.0
language:
  - en

Models trained from VITS-fast-fine-tuning

  • Three speakers: laoliang (老撁), specialweek, zhongli.
  • The model is trained on the C+J base model with 500 epochs.
  • Following the official instruction, we use a single long audio of laoliang (~5 minutes) with auxiliary data as training data.
  • After downloading models, you need to move finetune_speaker.json and G_latest.pth to /path/to/ VITS-fast-fine-tuning.
  • Finally, you can run your local gradio application via python VC_inference.py --model_dir ./G_latest.pth --share True

'''bash VITS-fast-fine-tuning β”œβ”€β”€β”€VC_inference.py β”œβ”€β”€β”€... β”œβ”€β”€β”€finetune_speaker.json └───G_latest.pth '''