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

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Scopus citations

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.

Original languageEnglish (US)
Title of host publicationSIGMORPHON 2021 - 18th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology, Proceedings of the Workshop
EditorsGarrett Nicolai, Kyle Gorman, Ryan Cotterell
PublisherAssociation for Computational Linguistics (ACL)
Pages126-130
Number of pages5
ISBN (Electronic)9781954085626
StatePublished - 2021
Event18th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology, SIGMORPHON 2021 - Virtual, Bangkok, Thailand
Duration: Aug 5 2021 → …

Publication series

NameSIGMORPHON 2021 - 18th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology, Proceedings of the Workshop

Conference

Conference18th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology, SIGMORPHON 2021
Country/TerritoryThailand
CityVirtual, Bangkok
Period8/5/21 → …

ASJC Scopus subject areas

  • Language and Linguistics
  • Artificial Intelligence
  • Human-Computer Interaction
  • Linguistics and Language

Fingerprint

Dive into the research topics of 'Data augmentation for low-resource grapheme-to-phoneme mapping'. Together they form a unique fingerprint.

Cite this