Back to Search Start Over

Modeling Prosodic Phrasing with Multi-Task Learning in Tacotron-based TTS

Authors :
Liu, Rui
Sisman, Berrak
Bao, Feilong
Gao, Guanglai
Li, Haizhou
Publication Year :
2020

Abstract

Tacotron-based end-to-end speech synthesis has shown remarkable voice quality. However, the rendering of prosody in the synthesized speech remains to be improved, especially for long sentences, where prosodic phrasing errors can occur frequently. In this paper, we extend the Tacotron-based speech synthesis framework to explicitly model the prosodic phrase breaks. We propose a multi-task learning scheme for Tacotron training, that optimizes the system to predict both Mel spectrum and phrase breaks. To our best knowledge, this is the first implementation of multi-task learning for Tacotron based TTS with a prosodic phrasing model. Experiments show that our proposed training scheme consistently improves the voice quality for both Chinese and Mongolian systems.<br />Comment: To appear in IEEE Signal Processing Letters (SPL)

Details

Database :
arXiv
Publication Type :
Report
Accession number :
edsarx.2008.05284
Document Type :
Working Paper
Full Text :
https://doi.org/10.1109/LSP.2020.3016564