A unified front-end framework for English text-to-speech synthesis

TL;DR

This paper introduces a unified front-end framework for English TTS systems that models the dependencies among modules, leading to improved performance across all components.

Contribution

It proposes a novel integrated framework for the TTS front-end that considers module interdependence, unlike previous isolated module approaches.

Findings

Achieves state-of-the-art performance in all front-end modules

Demonstrates the effectiveness of module dependency modeling

Improves overall TTS synthesis quality

Abstract

The front-end is a critical component of English text-to-speech (TTS) systems, responsible for extracting linguistic features that are essential for a text-to-speech model to synthesize speech, such as prosodies and phonemes. The English TTS front-end typically consists of a text normalization (TN) module, a prosody word prosody phrase (PWPP) module, and a grapheme-to-phoneme (G2P) module. However, current research on the English TTS front-end focuses solely on individual modules, neglecting the interdependence between them and resulting in sub-optimal performance for each module. Therefore, this paper proposes a unified front-end framework that captures the dependencies among the English TTS front-end modules. Extensive experiments have demonstrated that the proposed method achieves state-of-the-art (SOTA) performance in all modules.