Meta Learning Text-to-Speech Synthesis in over 7000 Languages
This work addresses the challenge of speech synthesis for underserved linguistic communities, enabling access to technology for languages lacking data, though it is incremental in combining existing techniques.
The authors tackled the problem of building a single text-to-speech synthesis system for over 7000 languages, many with insufficient data, and achieved zero-shot speech synthesis in languages without any available data through multilingual pretraining and meta learning.
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.