English

Incremental Text-to-Speech Synthesis Using Pseudo Lookahead with Large Pretrained Language Model

Sound 2021-05-26 v2 Audio and Speech Processing

Abstract

This letter presents an incremental text-to-speech (TTS) method that performs synthesis in small linguistic units while maintaining the naturalness of output speech. Incremental TTS is generally subject to a trade-off between latency and synthetic speech quality. It is challenging to produce high-quality speech with a low-latency setup that does not make much use of an unobserved future sentence (hereafter, "lookahead"). To resolve this issue, we propose an incremental TTS method that uses a pseudo lookahead generated with a language model to take the future contextual information into account without increasing latency. Our method can be regarded as imitating a human's incremental reading and uses pretrained GPT2, which accounts for the large-scale linguistic knowledge, for the lookahead generation. Evaluation results show that our method 1) achieves higher speech quality than the method taking only observed information into account and 2) achieves a speech quality equivalent to waiting for the future context observation.

Keywords

Cite

@article{arxiv.2012.12612,
  title  = {Incremental Text-to-Speech Synthesis Using Pseudo Lookahead with Large Pretrained Language Model},
  author = {Takaaki Saeki and Shinnosuke Takamichi and Hiroshi Saruwatari},
  journal= {arXiv preprint arXiv:2012.12612},
  year   = {2021}
}

Comments

Accepted for IEEE Signal Processing Letters

R2 v1 2026-06-23T21:16:54.739Z