Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers
Chengyi Wang, Sanyuan Chen, Yu Wu, Ziqiang Zhang, Long Zhou, Shujie Liu, Zhuo Chen, Yanqing Liu, Huaming Wang, Jinyu Li, Lei He, Sheng Zhao, Furu Wei
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/microsoft/unilmOfficialIn paperpytorch★ 22,060
- github.com/suno-ai/barkpytorch★ 39,059
- github.com/2noise/chatttspytorch★ 38,968
- github.com/plachtaa/vall-e-xpytorch★ 7,957
- github.com/serp-ai/bark-with-voice-clonepytorch★ 3,340
- github.com/enhuiz/vall-epytorch★ 2,992
- github.com/lifeiteng/vall-epytorch★ 2,211
Abstract
We introduce a language modeling approach for text to speech synthesis (TTS). Specifically, we train a neural codec language model (called Vall-E) using discrete codes derived from an off-the-shelf neural audio codec model, and regard TTS as a conditional language modeling task rather than continuous signal regression as in previous work. During the pre-training stage, we scale up the TTS training data to 60K hours of English speech which is hundreds of times larger than existing systems. Vall-E emerges in-context learning capabilities and can be used to synthesize high-quality personalized speech with only a 3-second enrolled recording of an unseen speaker as an acoustic prompt. Experiment results show that Vall-E significantly outperforms the state-of-the-art zero-shot TTS system in terms of speech naturalness and speaker similarity. In addition, we find Vall-E could preserve the speaker's emotion and acoustic environment of the acoustic prompt in synthesis. See https://aka.ms/valle for demos of our work.