ByT5 model for massively multilingual grapheme-to-phoneme conversion
This addresses the problem of improving grapheme-to-phoneme conversion for multilingual applications, particularly benefiting low-resource languages, though it is incremental as it builds on existing ByT5 methods.
The study tackled massively multilingual grapheme-to-phoneme conversion by training models based on ByT5 on a dataset covering around 100 languages, finding that ByT5 outperformed token-based mT5 and lowered phone error rates through joint learning across languages.
In this study, we tackle massively multilingual grapheme-to-phoneme conversion through implementing G2P models based on ByT5. We have curated a G2P dataset from various sources that covers around 100 languages and trained large-scale multilingual G2P models based on ByT5. We found that ByT5 operating on byte-level inputs significantly outperformed the token-based mT5 model in terms of multilingual G2P. Pairwise comparison with monolingual models in these languages suggests that multilingual ByT5 models generally lower the phone error rate by jointly learning from a variety of languages. The pretrained model can further benefit low resource G2P through zero-shot prediction on unseen languages or provides pretrained weights for finetuning, which helps the model converge to a lower phone error rate than randomly initialized weights. To facilitate future research on multilingual G2P, we make available our code and pretrained multilingual G2P models at: https://github.com/lingjzhu/CharsiuG2P.