Meta Learning Text-to-Speech Synthesis in over 7000 Languages

TL;DR

This paper introduces a universal text-to-speech system capable of synthesizing speech in over 7000 languages, including many with no available data, using multilingual pretraining and meta learning.

Contribution

It presents a novel approach combining multilingual pretraining and meta learning to enable zero-shot TTS in extremely low-resource languages.

Findings

Effective zero-shot synthesis demonstrated across diverse languages

System outperforms baseline models in objective and human evaluations

Public release of code and models to support linguistic diversity

Abstract

In this work, we take on the challenging task of building a single text-to-speech synthesis system that is capable of generating speech in over 7000 languages, many of which lack sufficient data for traditional TTS development. By leveraging a novel integration of massively multilingual pretraining and meta learning to approximate language representations, our approach enables zero-shot speech synthesis in languages without any available data. We validate our system's performance through objective measures and human evaluation across a diverse linguistic landscape. By releasing our code and models publicly, we aim to empower communities with limited linguistic resources and foster further innovation in the field of speech technology.