English

A unified sequence-to-sequence front-end model for Mandarin text-to-speech synthesis

Computation and Language 2019-11-12 v1 Sound Audio and Speech Processing

Abstract

In Mandarin text-to-speech (TTS) system, the front-end text processing module significantly influences the intelligibility and naturalness of synthesized speech. Building a typical pipeline-based front-end which consists of multiple individual components requires extensive efforts. In this paper, we proposed a unified sequence-to-sequence front-end model for Mandarin TTS that converts raw texts to linguistic features directly. Compared to the pipeline-based front-end, our unified front-end can achieve comparable performance in polyphone disambiguation and prosody word prediction, and improve intonation phrase prediction by 0.0738 in F1 score. We also implemented the unified front-end with Tacotron and WaveRNN to build a Mandarin TTS system. The synthesized speech by that got a comparable MOS (4.38) with the pipeline-based front-end (4.37) and close to human recordings (4.49).

Keywords

Cite

@article{arxiv.1911.04111,
  title  = {A unified sequence-to-sequence front-end model for Mandarin text-to-speech synthesis},
  author = {Junjie Pan and Xiang Yin and Zhiling Zhang and Shichao Liu and Yang Zhang and Zejun Ma and Yuxuan Wang},
  journal= {arXiv preprint arXiv:1911.04111},
  year   = {2019}
}

Comments

Submitted to ICASSP 2020

R2 v1 2026-06-23T12:11:13.453Z