Easily train and convert voices using a retrieval-based model, even with minimal source audio.
It's a Python-based web UI that lets users quickly train a voice conversion model, often with less than 10 minutes of voice data. The framework uses a VITS-based architecture and includes a pre-trained base model built on 50 hours of open-source VCTK data. You'll find a user-friendly interface for both training models and performing real-time voice changes.
Easily train and convert voices using a retrieval-based model, even with minimal source audio.
Anyone wanting to create custom AI voices for singing, speaking, or character work with a straightforward workflow.
Not enough data yet. Star history will appear after a few days of tracking.
