Grapheme-Based Cross-Language Forced Alignment: Results with Uralic Languages

Juho Leinonen; Sami Virpioja; Mikko Kurimo

Grapheme-Based Cross-Language Forced Alignment: Results with Uralic Languages

Juho Leinonen, Sami Virpioja, Mikko Kurimo

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

Abstract

Forced alignment is an effective process to speed up linguistic research. However, most forced aligners are language-dependent, and under-resourced languages rarely have enough resources to train an acoustic model for an aligner. We present a new Finnish grapheme-based forced aligner and demonstrate its performance by aligning multiple Uralic languages and English as an unrelated language. We show that even a simple non-expert created grapheme-to-phoneme mapping can result in useful word alignments.

Original language	English
Title of host publication	Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa)
Editors	Simon Dobnik, Lilja Øvrelid
Number of pages	6
Place of Publication	Linköping
Publisher	Linköping University Electronic Press
Publication date	1 May 2021
Pages	345-350
ISBN (Electronic)	978-91-7929-614-8
Publication status	Published - 1 May 2021
MoE publication type	A4 Article in conference proceedings
Event	Nordic Conference on Computational Linguistics - [Online event], Reykjavik, Iceland Duration: 31 May 2021 → 2 Jun 2021 Conference number: 23 https://nodalida2021.github.io/index.html

Publication series

Name	Linköping Electronic Conference Proceedings
Publisher	Linköping University Electronic Press
Number	78
ISSN (Print)	1650-3686
ISSN (Electronic)	1650-3740

Name	NEALT Proceedings Series
Publisher	University of Tartu
Number	45
ISSN (Print)	1736-8197
ISSN (Electronic)	1736-6305

Fields of Science

113 Computer and information sciences
6121 Languages

Access to Document

2021.nodalida-main_362Final published version, 259 KBLicence: CC BY

https://www.aclweb.org/anthology/2021.nodalida-main.36.pdfLicence: CC BY

Cite this

Leinonen, J., Virpioja, S., & Kurimo, M. (2021). Grapheme-Based Cross-Language Forced Alignment: Results with Uralic Languages. In S. Dobnik, & L. Øvrelid (Eds.), Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa) (pp. 345-350). (Linköping Electronic Conference Proceedings; No. 78), (NEALT Proceedings Series; No. 45). Linköping University Electronic Press. https://www.aclweb.org/anthology/2021.nodalida-main.36.pdf

Leinonen, Juho ; Virpioja, Sami ; Kurimo, Mikko. / Grapheme-Based Cross-Language Forced Alignment : Results with Uralic Languages. Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa). editor / Simon Dobnik ; Lilja Øvrelid. Linköping : Linköping University Electronic Press, 2021. pp. 345-350 (Linköping Electronic Conference Proceedings; 78). (NEALT Proceedings Series; 45).

@inproceedings{bf79c4fac97543098a05b6d082ef327a,

title = "Grapheme-Based Cross-Language Forced Alignment: Results with Uralic Languages",

abstract = "Forced alignment is an effective process to speed up linguistic research. However, most forced aligners are language-dependent, and under-resourced languages rarely have enough resources to train an acoustic model for an aligner. We present a new Finnish grapheme-based forced aligner and demonstrate its performance by aligning multiple Uralic languages and English as an unrelated language. We show that even a simple non-expert created grapheme-to-phoneme mapping can result in useful word alignments.",

keywords = "113 Computer and information sciences, 6121 Languages",

author = "Juho Leinonen and Sami Virpioja and Mikko Kurimo",

year = "2021",

month = may,

day = "1",

language = "English",

series = "Link{\"o}ping Electronic Conference Proceedings",

publisher = "Link{\"o}ping University Electronic Press",

number = "78",

pages = "345--350",

editor = "Dobnik, {Simon } and {\O}vrelid, {Lilja }",

booktitle = "Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa)",

address = "Sweden",

note = "Nordic Conference on Computational Linguistics, NoDaLiDa ; Conference date: 31-05-2021 Through 02-06-2021",

url = "https://nodalida2021.github.io/index.html",

}

Leinonen, J, Virpioja, S & Kurimo, M 2021, Grapheme-Based Cross-Language Forced Alignment: Results with Uralic Languages. in S Dobnik & L Øvrelid (eds), Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa). Linköping Electronic Conference Proceedings, no. 78, NEALT Proceedings Series, no. 45, Linköping University Electronic Press, Linköping, pp. 345-350, Nordic Conference on Computational Linguistics, Reykjavik, Iceland, 31/05/2021. <https://www.aclweb.org/anthology/2021.nodalida-main.36.pdf>

Grapheme-Based Cross-Language Forced Alignment : Results with Uralic Languages. / Leinonen, Juho; Virpioja, Sami; Kurimo, Mikko.

Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa). ed. / Simon Dobnik; Lilja Øvrelid. Linköping : Linköping University Electronic Press, 2021. p. 345-350 (Linköping Electronic Conference Proceedings; No. 78), (NEALT Proceedings Series; No. 45).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Grapheme-Based Cross-Language Forced Alignment

T2 - Nordic Conference on Computational Linguistics

AU - Leinonen, Juho

AU - Virpioja, Sami

AU - Kurimo, Mikko

N1 - Conference code: 23

PY - 2021/5/1

Y1 - 2021/5/1

N2 - Forced alignment is an effective process to speed up linguistic research. However, most forced aligners are language-dependent, and under-resourced languages rarely have enough resources to train an acoustic model for an aligner. We present a new Finnish grapheme-based forced aligner and demonstrate its performance by aligning multiple Uralic languages and English as an unrelated language. We show that even a simple non-expert created grapheme-to-phoneme mapping can result in useful word alignments.

AB - Forced alignment is an effective process to speed up linguistic research. However, most forced aligners are language-dependent, and under-resourced languages rarely have enough resources to train an acoustic model for an aligner. We present a new Finnish grapheme-based forced aligner and demonstrate its performance by aligning multiple Uralic languages and English as an unrelated language. We show that even a simple non-expert created grapheme-to-phoneme mapping can result in useful word alignments.

KW - 113 Computer and information sciences

KW - 6121 Languages

UR - https://www.aclweb.org/anthology/2021.nodalida-main.36

M3 - Conference contribution

T3 - Linköping Electronic Conference Proceedings

SP - 345

EP - 350

BT - Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa)

A2 - Dobnik, Simon

A2 - Øvrelid, Lilja

PB - Linköping University Electronic Press

CY - Linköping

Y2 - 31 May 2021 through 2 June 2021

ER -

Leinonen J, Virpioja S, Kurimo M. Grapheme-Based Cross-Language Forced Alignment: Results with Uralic Languages. In Dobnik S, Øvrelid L, editors, Proceedings of the 23rd Nordic Conference on Computational Linguistics (NoDaLiDa). Linköping: Linköping University Electronic Press. 2021. p. 345-350. (Linköping Electronic Conference Proceedings; 78). (NEALT Proceedings Series; 45).

Grapheme-Based Cross-Language Forced Alignment: Results with Uralic Languages

Abstract

Publication series

Fields of Science

Access to Document

Other files and links

Cite this