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

Rui Liu, Berrak Sisman, Feilong Bao, Guang Lai Gao, Haizhou Li
2020 IEEE Signal Processing Letters  
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
more » ... s. 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.
doi:10.1109/lsp.2020.3016564 fatcat:q7rd6md5mnbrtpsyjpbaoiv5ou