Zero-shot Learning for Grapheme to Phoneme Conversion with Language Ensemble

Xinjian Li, Florian Metze, David Mortensen, Shinji Watanabe, Alan Black


Abstract
Grapheme-to-Phoneme (G2P) has many applications in NLP and speech fields. Most existing work focuses heavily on languages with abundant training datasets, which limits the scope of target languages to less than 100 languages. This work attempts to apply zero-shot learning to approximate G2P models for all low-resource and endangered languages in Glottolog (about 8k languages). For any unseen target language, we first build the phylogenetic tree (i.e. language family tree) to identify top-k nearest languages for which we have training sets. Then we run models of those languages to obtain a hypothesis set, which we combine into a confusion network to propose a most likely hypothesis as an approximation to the target language. We test our approach on over 600 unseen languages and demonstrate it significantly outperforms baselines.
Anthology ID:
2022.findings-acl.166
Volume:
Findings of the Association for Computational Linguistics: ACL 2022
Month:
May
Year:
2022
Address:
Dublin, Ireland
Editors:
Smaranda Muresan, Preslav Nakov, Aline Villavicencio
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2106–2115
Language:
URL:
https://aclanthology.org/2022.findings-acl.166
DOI:
10.18653/v1/2022.findings-acl.166
Bibkey:
Cite (ACL):
Xinjian Li, Florian Metze, David Mortensen, Shinji Watanabe, and Alan Black. 2022. Zero-shot Learning for Grapheme to Phoneme Conversion with Language Ensemble. In Findings of the Association for Computational Linguistics: ACL 2022, pages 2106–2115, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
Zero-shot Learning for Grapheme to Phoneme Conversion with Language Ensemble (Li et al., Findings 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.findings-acl.166.pdf
Video:
 https://aclanthology.org/2022.findings-acl.166.mp4
Code
 xinjli/transphone