Abstract
In this paper we explore a very simple neural approach to mapping orthography to phonetic transcription in a low-resource context. The basic idea is to start from a baseline system and focus all efforts on data augmentation. We will see that some techniques work, but others do not.- Anthology ID:
- 2021.sigmorphon-1.14
- Volume:
- Proceedings of the 18th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology
- Month:
- August
- Year:
- 2021
- Address:
- Online
- Editors:
- Garrett Nicolai, Kyle Gorman, Ryan Cotterell
- Venue:
- SIGMORPHON
- SIG:
- SIGMORPHON
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 126–130
- Language:
- URL:
- https://aclanthology.org/2021.sigmorphon-1.14
- DOI:
- 10.18653/v1/2021.sigmorphon-1.14
- Cite (ACL):
- Michael Hammond. 2021. Data augmentation for low-resource grapheme-to-phoneme mapping. In Proceedings of the 18th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology, pages 126–130, Online. Association for Computational Linguistics.
- Cite (Informal):
- Data augmentation for low-resource grapheme-to-phoneme mapping (Hammond, SIGMORPHON 2021)
- PDF:
- https://preview.aclanthology.org/ml4al-ingestion/2021.sigmorphon-1.14.pdf