Skip to content
conftrace_
2021 ACL ACL 2021

Data augmentation for low-resource grapheme-to-phoneme mapping

Abstract

AbstractIn 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.

🌉 Interdisciplinary Bridge - Artificial Intelligence and Natural Language Processing and Speech & Audio

Authors

Michael Hammond

Topics

Machine Learning > Application Areas > Data Augmentation Natural Language Processing > Generation > Language Modeling Natural Language Processing > Applications > Machine Translation Speech & Audio > Recognition > Speech Recognition

Keywords

data augmentation low-resource setting grapheme-to-phoneme conversion neural approach
Download PDF

More from ACL 2021

All ACL 2021 papers →
Unsupervised Paradigm Clustering Using Transformation Rules
Paradigm Clustering with Weighted Edit Distance
Results of the Second SIGMORPHON Shared Task on Multilingual Grapheme-to-Phoneme Conversion
Linguistic Knowledge in Multilingual Grapheme-to-Phoneme Conversion
Avengers, Ensemble! Benefits of ensembling in grapheme-to-phoneme prediction
CLUZH at SIGMORPHON 2021 Shared Task on Multilingual Grapheme-to-Phoneme Conversion: Variations on a Baseline