Low-resource grapheme-to-phoneme mapping with phonetically-conditioned transfer

Michael Hammond

doi:10.18653/v1/2023.sigmorphon-1.29

Low-resource grapheme-to-phoneme mapping with phonetically-conditioned transfer

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

In this paper we explore a very simple nonneural approach to mapping orthography to phonetic transcription in a low-resource context with transfer data from a related language. We start from a baseline system and focus our efforts on data augmentation. We make three principal moves. First, we start with an HMMbased system (Novak et al., 2012). Second, we augment our basic system by recombining legal substrings in restricted fashion (Ryan and Hulden, 2020). Finally, we limit our transfer data by only using training pairs where the phonetic form shares all bigrams with the target language.

Original language	English (US)
Title of host publication	ACL 2023 - 20th SIGMORPHON Workshop on Computational Morphology, Phonology, and Phonetics, CMPP 2023
Editors	Garrett Nicolai, Eleanor Chodroff, Cagri Coltekin, Fred Mailhot
Publisher	Association for Computational Linguistics (ACL)
Pages	245-248
Number of pages	4
ISBN (Electronic)	9781959429937
DOIs	https://doi.org/10.18653/v1/2023.sigmorphon-1.29
State	Published - 2023
Event	20th SIGMORPHON Workshop on Computational Morphology, Phonology, and Phonetics, CMPP 2023, as part of ACL 2023 - Toronto, Canada Duration: Jul 14 2023 → …

Publication series

Name	Proceedings of the Annual Meeting of the Association for Computational Linguistics
ISSN (Print)	0736-587X

Conference

Conference	20th SIGMORPHON Workshop on Computational Morphology, Phonology, and Phonetics, CMPP 2023, as part of ACL 2023
Country/Territory	Canada
City	Toronto
Period	7/14/23 → …

ASJC Scopus subject areas

Computer Science Applications
Linguistics and Language
Language and Linguistics

Access to Document

10.18653/v1/2023.sigmorphon-1.29

Cite this

Hammond, M. (2023). Low-resource grapheme-to-phoneme mapping with phonetically-conditioned transfer. In G. Nicolai, E. Chodroff, C. Coltekin, & F. Mailhot (Eds.), ACL 2023 - 20th SIGMORPHON Workshop on Computational Morphology, Phonology, and Phonetics, CMPP 2023 (pp. 245-248). (Proceedings of the Annual Meeting of the Association for Computational Linguistics). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/2023.sigmorphon-1.29

Low-resource grapheme-to-phoneme mapping with phonetically-conditioned transfer. / Hammond, Michael.
ACL 2023 - 20th SIGMORPHON Workshop on Computational Morphology, Phonology, and Phonetics, CMPP 2023. ed. / Garrett Nicolai; Eleanor Chodroff; Cagri Coltekin; Fred Mailhot. Association for Computational Linguistics (ACL), 2023. p. 245-248 (Proceedings of the Annual Meeting of the Association for Computational Linguistics).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Hammond, M 2023, Low-resource grapheme-to-phoneme mapping with phonetically-conditioned transfer. in G Nicolai, E Chodroff, C Coltekin & F Mailhot (eds), ACL 2023 - 20th SIGMORPHON Workshop on Computational Morphology, Phonology, and Phonetics, CMPP 2023. Proceedings of the Annual Meeting of the Association for Computational Linguistics, Association for Computational Linguistics (ACL), pp. 245-248, 20th SIGMORPHON Workshop on Computational Morphology, Phonology, and Phonetics, CMPP 2023, as part of ACL 2023, Toronto, Canada, 7/14/23. https://doi.org/10.18653/v1/2023.sigmorphon-1.29

Hammond M. Low-resource grapheme-to-phoneme mapping with phonetically-conditioned transfer. In Nicolai G, Chodroff E, Coltekin C, Mailhot F, editors, ACL 2023 - 20th SIGMORPHON Workshop on Computational Morphology, Phonology, and Phonetics, CMPP 2023. Association for Computational Linguistics (ACL). 2023. p. 245-248. (Proceedings of the Annual Meeting of the Association for Computational Linguistics). doi: 10.18653/v1/2023.sigmorphon-1.29

Hammond, Michael. / Low-resource grapheme-to-phoneme mapping with phonetically-conditioned transfer. ACL 2023 - 20th SIGMORPHON Workshop on Computational Morphology, Phonology, and Phonetics, CMPP 2023. editor / Garrett Nicolai ; Eleanor Chodroff ; Cagri Coltekin ; Fred Mailhot. Association for Computational Linguistics (ACL), 2023. pp. 245-248 (Proceedings of the Annual Meeting of the Association for Computational Linguistics).

@inproceedings{63f52759d7ce49dd86f4631680454a88,

title = "Low-resource grapheme-to-phoneme mapping with phonetically-conditioned transfer",

abstract = "In this paper we explore a very simple nonneural approach to mapping orthography to phonetic transcription in a low-resource context with transfer data from a related language. We start from a baseline system and focus our efforts on data augmentation. We make three principal moves. First, we start with an HMMbased system (Novak et al., 2012). Second, we augment our basic system by recombining legal substrings in restricted fashion (Ryan and Hulden, 2020). Finally, we limit our transfer data by only using training pairs where the phonetic form shares all bigrams with the target language.",

author = "Michael Hammond",

note = "Publisher Copyright: {\textcopyright} 2023 Association for Computational Linguistics.; 20th SIGMORPHON Workshop on Computational Morphology, Phonology, and Phonetics, CMPP 2023, as part of ACL 2023 ; Conference date: 14-07-2023",

year = "2023",

doi = "10.18653/v1/2023.sigmorphon-1.29",

language = "English (US)",

series = "Proceedings of the Annual Meeting of the Association for Computational Linguistics",

publisher = "Association for Computational Linguistics (ACL)",

pages = "245--248",

editor = "Garrett Nicolai and Eleanor Chodroff and Cagri Coltekin and Fred Mailhot",

booktitle = "ACL 2023 - 20th SIGMORPHON Workshop on Computational Morphology, Phonology, and Phonetics, CMPP 2023",

}

TY - GEN

T1 - Low-resource grapheme-to-phoneme mapping with phonetically-conditioned transfer

AU - Hammond, Michael

PY - 2023

Y1 - 2023

N2 - In this paper we explore a very simple nonneural approach to mapping orthography to phonetic transcription in a low-resource context with transfer data from a related language. We start from a baseline system and focus our efforts on data augmentation. We make three principal moves. First, we start with an HMMbased system (Novak et al., 2012). Second, we augment our basic system by recombining legal substrings in restricted fashion (Ryan and Hulden, 2020). Finally, we limit our transfer data by only using training pairs where the phonetic form shares all bigrams with the target language.

AB - In this paper we explore a very simple nonneural approach to mapping orthography to phonetic transcription in a low-resource context with transfer data from a related language. We start from a baseline system and focus our efforts on data augmentation. We make three principal moves. First, we start with an HMMbased system (Novak et al., 2012). Second, we augment our basic system by recombining legal substrings in restricted fashion (Ryan and Hulden, 2020). Finally, we limit our transfer data by only using training pairs where the phonetic form shares all bigrams with the target language.

UR - http://www.scopus.com/inward/record.url?scp=85175400254&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85175400254&partnerID=8YFLogxK

U2 - 10.18653/v1/2023.sigmorphon-1.29

DO - 10.18653/v1/2023.sigmorphon-1.29

M3 - Conference contribution

AN - SCOPUS:85175400254

T3 - Proceedings of the Annual Meeting of the Association for Computational Linguistics

SP - 245

EP - 248

BT - ACL 2023 - 20th SIGMORPHON Workshop on Computational Morphology, Phonology, and Phonetics, CMPP 2023

A2 - Nicolai, Garrett

A2 - Chodroff, Eleanor

A2 - Coltekin, Cagri

A2 - Mailhot, Fred

PB - Association for Computational Linguistics (ACL)

T2 - 20th SIGMORPHON Workshop on Computational Morphology, Phonology, and Phonetics, CMPP 2023, as part of ACL 2023

Y2 - 14 July 2023

ER -

Low-resource grapheme-to-phoneme mapping with phonetically-conditioned transfer

Abstract

Publication series

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this