Exploiting Syntactic Features in a Parsed Tree to Improve End-to-End TTS
This work addresses pronunciation and prosody issues in text-to-speech synthesis for applications requiring natural-sounding speech, but it is incremental as it builds on existing end-to-end TTS methods.
The paper tackled the problem of limited acoustic/phonetic coverage in end-to-end TTS by exploiting syntactic features from parsed trees, such as phrase structure and word relations, to improve speech quality. Experimental results on three test sets showed effectiveness in enhancing pronunciation clarity, prosody, and naturalness over a baseline system.
The end-to-end TTS, which can predict speech directly from a given sequence of graphemes or phonemes, has shown improved performance over the conventional TTS. However, its predicting capability is still limited by the acoustic/phonetic coverage of the training data, usually constrained by the training set size. To further improve the TTS quality in pronunciation, prosody and perceived naturalness, we propose to exploit the information embedded in a syntactically parsed tree where the inter-phrase/word information of a sentence is organized in a multilevel tree structure. Specifically, two key features: phrase structure and relations between adjacent words are investigated. Experimental results in subjective listening, measured on three test sets, show that the proposed approach is effective to improve the pronunciation clarity, prosody and naturalness of the synthesized speech of the baseline system.