GraphSpeech: Syntax-Aware Graph Attention Network For Neural Speech Synthesis
This is an incremental improvement for neural speech synthesis, addressing syntactic modeling to enhance output quality.
The authors tackled the problem of Transformer-based text-to-speech synthesis lacking syntactic sentence-level associations by proposing GraphSpeech, a syntax-aware graph attention network, which consistently outperformed the baseline in spectrum and prosody rendering.
Attention-based end-to-end text-to-speech synthesis (TTS) is superior to conventional statistical methods in many ways. Transformer-based TTS is one of such successful implementations. While Transformer TTS models the speech frame sequence well with a self-attention mechanism, it does not associate input text with output utterances from a syntactic point of view at sentence level. We propose a novel neural TTS model, denoted as GraphSpeech, that is formulated under graph neural network framework. GraphSpeech encodes explicitly the syntactic relation of input lexical tokens in a sentence, and incorporates such information to derive syntactically motivated character embeddings for TTS attention mechanism. Experiments show that GraphSpeech consistently outperforms the Transformer TTS baseline in terms of spectrum and prosody rendering of utterances.