Text-aware and Context-aware Expressive Audiobook Speech Synthesis

TL;DR

This paper introduces a novel text-aware and context-aware style modeling approach for expressive audiobook speech synthesis, capturing diverse narration styles without manual labels, thereby enhancing naturalness and expressiveness.

Contribution

It proposes a new style modeling framework that combines contrastive learning and cross-sentence context encoding, integrated into existing TTS models for improved audiobook narration.

Findings

Effectively captures diverse narration styles.

Improves naturalness and expressiveness of synthesized speech.

Enhances coherence across sentences in audiobook synthesis.

Abstract

Recent advances in text-to-speech have significantly improved the expressiveness of synthetic speech. However, a major challenge remains in generating speech that captures the diverse styles exhibited by professional narrators in audiobooks without relying on manually labeled data or reference speech. To address this problem, we propose a text-aware and context-aware(TACA) style modeling approach for expressive audiobook speech synthesis. We first establish a text-aware style space to cover diverse styles via contrastive learning with the supervision of the speech style. Meanwhile, we adopt a context encoder to incorporate cross-sentence information and the style embedding obtained from text. Finally, we introduce the context encoder to two typical TTS models, VITS-based TTS and language model-based TTS. Experimental results demonstrate that our proposed approach can effectively capture diverse styles and coherent prosody, and consequently improves naturalness and expressiveness in audiobook speech synthesis.