Learning to Speak from Text: Zero-Shot Multilingual Text-to-Speech with Unsupervised Text Pretraining
Learning to Speak from Text: Zero-Shot Multilingual Text-to-Speech with Unsupervised Text Pretraining
Takaaki Saeki, Soumi Maiti, Xinjian Li, Shinji Watanabe, Shinnosuke Takamichi, Hiroshi Saruwatari
Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence
Main Track. Pages 5179-5187.
https://doi.org/10.24963/ijcai.2023/575
While neural text-to-speech (TTS) has achieved human-like natural synthetic speech, multilingual TTS systems are limited to resource-rich languages due to the need for paired text and studio-quality audio data. This paper proposes a method for zero-shot multilingual TTS using text-only data for the target language. The use of text-only data allows the development of TTS systems for low-resource languages for which only textual resources are available, making TTS accessible to thousands of languages. Inspired by the strong cross-lingual transferability of multilingual language models, our framework first performs masked language model pretraining with multilingual text-only data. Then we train this model with a paired data in a supervised manner, while freezing a language-aware embedding layer. This allows inference even for languages not included in the paired data but present in the text-only data. Evaluation results demonstrate highly intelligible zero-shot TTS with a character error rate of less than 12% for an unseen language.
Keywords:
Natural Language Processing: NLP: Speech
Natural Language Processing: NLP: Language models