MultiVerse: Efficient and Expressive Zero-Shot Multi-Task Text-to-Speech

TL;DR

MultiVerse is a zero-shot multi-task TTS system that efficiently performs speech synthesis and style transfer with minimal data, leveraging disentanglement and advanced prosody modeling to achieve high-quality, cross-lingual, and style-consistent speech output.

Contribution

It introduces a novel zero-shot multi-task TTS framework that requires less data, employs source-filter disentanglement, and combines autoregressive and non-autoregressive prosody modeling for improved performance.

Findings

Achieves zero-shot TTS performance comparable to data-driven systems with less data.

Significantly outperforms other zero-shot TTS systems trained on the same small datasets.

Enhances prosody similarity through a novel combined prosody modeling approach.

Abstract

Text-to-speech (TTS) systems that scale up the amount of training data have achieved significant improvements in zero-shot speech synthesis. However, these systems have certain limitations: they require a large amount of training data, which increases costs, and often overlook prosody similarity. To address these issues, we propose MultiVerse, a zero-shot multi-task TTS system that is able to perform TTS or speech style transfer in zero-shot and cross-lingual conditions. MultiVerse requires much less training data than traditional data-driven approaches. To ensure zero-shot performance even with limited data, we leverage source-filter theory-based disentanglement, utilizing the prompt for modeling filter-related and source-related representations. Additionally, to further enhance prosody similarity, we adopt a prosody modeling approach combining prompt-based autoregressive and non-autoregressive methods. Evaluations demonstrate the remarkable zero-shot multi-task TTS performance of MultiVerse and show that MultiVerse not only achieves zero-shot TTS performance comparable to data-driven TTS systems with much less data, but also significantly outperforms other zero-shot TTS systems trained with the same small amount of data. In particular, our novel prosody modeling technique significantly contributes to MultiVerse's ability to generate speech with high prosody similarity to the given prompts. Our samples are available at https://nc-ai.github.io/speech/publications/multiverse/index.html