Plachtaa/VALL-E-X
An open-source reproduction of Microsoft's VALL-E X zero-shot multilingual text-to-speech synthesis model with voice cloning capabilities.
Not currently ranked — collecting fresh signals.
star history
This repository provides a trained implementation of VALL-E X, a zero-shot TTS model capable of synthesizing speech in multiple languages from just a 3-second enrollment recording, preserving the speaker’s voice characteristics and emotional tone. The model uses a GPT-style autoregressive decoder with an Encodec or Vocos neural codec decoder for audio generation.
Frequently asked
- What is Plachtaa/VALL-E-X?
- An open-source reproduction of Microsoft's VALL-E X zero-shot multilingual text-to-speech synthesis model with voice cloning capabilities.
- Is VALL-E-X open source?
- Yes — Plachtaa/VALL-E-X is open source, released under the MIT license.
- What language is VALL-E-X written in?
- Plachtaa/VALL-E-X is primarily written in Python.
- How popular is VALL-E-X?
- Plachtaa/VALL-E-X has 7.9k stars on GitHub.
- Where can I find VALL-E-X?
- Plachtaa/VALL-E-X is on GitHub at https://github.com/Plachtaa/VALL-E-X.