Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers
Paper • 2301.02111 • Published • 1
VALL-E
Pre-trained checkpoint of Vall-E with AudioDec trained on WenetSpeech4TTS.
Checkpoints
We provide three checkpoints trained/fine-tuned on different subset of WenetSpeech4TTS.
- VALL-E Basic : VALL-E trained with the WenetSpeech4TTS Basic subset
- VALL-E Standard : VALL-E Basic fine-tuning with the WenetSpeech4TTS Standard subset
- VALL-E Premium : VALL-E Standard fine-tuning with the WenetSpeech4TTS Premium subset
Useage
Inference code and more details : dukGuo/valle-audiodec.
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support