Zero-Shot Text-to-Speech for Vietnamese

TL;DR

This paper presents PhoAudiobook, a large Vietnamese TTS dataset, and evaluates three zero-shot models, showing improved performance and robustness in synthesizing diverse linguistic content.

Contribution

Introduction of PhoAudiobook, a comprehensive Vietnamese TTS dataset, and an empirical evaluation of zero-shot TTS models demonstrating their effectiveness.

Findings

PhoAudiobook improves model performance across metrics.

VALL-E and VoiceCraft excel in short sentence synthesis.

PhoAudiobook is publicly available for research.

Abstract

This paper introduces PhoAudiobook, a newly curated dataset comprising 941 hours of high-quality audio for Vietnamese text-to-speech. Using PhoAudiobook, we conduct experiments on three leading zero-shot TTS models: VALL-E, VoiceCraft, and XTTS-V2. Our findings demonstrate that PhoAudiobook consistently enhances model performance across various metrics. Moreover, VALL-E and VoiceCraft exhibit superior performance in synthesizing short sentences, highlighting their robustness in handling diverse linguistic contexts. We publicly release PhoAudiobook to facilitate further research and development in Vietnamese text-to-speech.