Improving code-switched ASR with linguistic information

Jie Chi; Peter Bell

Improving code-switched ASR with linguistic information

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

Abstract

This paper seeks to improve the performance of automatic speech recognition (ASR) systems operating on code-switched speech. Code-switching refers to the alternation of languages within a conversation, a phenomenon that is of increasing importance considering the rapid rise in the number of bilingual speakers in the world. It is particularly challenging for ASR owing to the relative scarcity of code-switching speech and text data, even when the individual languages are themselves well-resourced. This paper proposes to overcome this challenge by applying linguistic theories in order to generate more realistic code-switching text, necessary for language modelling in ASR. Working with English-Spanish code-switching, we find that Equivalence Constraint theory and part-of-speech labelling are particularly helpful for text generation, and bring 2% improvement to ASR performance.

Original language	English
Title of host publication	Proceedings of the 29th International Conference on Computational Linguistics
Editors	Nicoletta Calzolari, Chu-Ren Huang, Hansaem Kim, James Pustejovsky, Leo Wanner, Key-Sun Choi, Pum-Mo Ryu, Hsin-Hsi Chen, Lucia Donatelli, Heng Ji, Sadao Kurohashi, Patrizia Paggio, Nianwen Xue, Seokhwan Kim, Younggyun Hahm, Zhong He, Tony Kyungil Lee, Enrico Santus, Francis Bond, Seung-Hoon Na
Publisher	ACL Anthology
Pages	7171-7176
Number of pages	6
Volume	29
Publication status	Published - 3 Nov 2022
Event	The 29th International Conference on Computational Linguistics, 2022 - Gyeongju, Korea, Republic of Duration: 12 Oct 2022 → 17 Oct 2022 Conference number: 29 https://coling2022.org/

Publication series

Name	COLING
Publisher	ACL Anthology
Number	1
Volume	29
ISSN (Electronic)	2591-2093

Conference

Conference	The 29th International Conference on Computational Linguistics, 2022
Abbreviated title	COLING 2022
Country/Territory	Korea, Republic of
City	Gyeongju
Period	12/10/22 → 17/10/22
Internet address	https://coling2022.org/

Access to Document

Improving code_CHI_DOA15082022_AFVAccepted author manuscript, 132 KBLicence: Creative Commons: Attribution (CC-BY)
Improving code_CHI_DOA15082022_VOR_CC-BYFinal published version, 199 KBLicence: Creative Commons: Attribution (CC-BY)

https://aclanthology.org/2022.coling-1.627/Licence: Creative Commons: Attribution (CC-BY)

Cite this

Chi, J., & Bell, P. (2022). Improving code-switched ASR with linguistic information. In N. Calzolari, C.-R. Huang, H. Kim, J. Pustejovsky, L. Wanner, K.-S. Choi, P.-M. Ryu, H.-H. Chen, L. Donatelli, H. Ji, S. Kurohashi, P. Paggio, N. Xue, S. Kim, Y. Hahm, Z. He, T. K. Lee, E. Santus, F. Bond, & S.-H. Na (Eds.), Proceedings of the 29th International Conference on Computational Linguistics (Vol. 29, pp. 7171-7176). (COLING; Vol. 29, No. 1). ACL Anthology. https://aclanthology.org/2022.coling-1.627/

Chi, Jie ; Bell, Peter. / Improving code-switched ASR with linguistic information. Proceedings of the 29th International Conference on Computational Linguistics. editor / Nicoletta Calzolari ; Chu-Ren Huang ; Hansaem Kim ; James Pustejovsky ; Leo Wanner ; Key-Sun Choi ; Pum-Mo Ryu ; Hsin-Hsi Chen ; Lucia Donatelli ; Heng Ji ; Sadao Kurohashi ; Patrizia Paggio ; Nianwen Xue ; Seokhwan Kim ; Younggyun Hahm ; Zhong He ; Tony Kyungil Lee ; Enrico Santus ; Francis Bond ; Seung-Hoon Na. Vol. 29 ACL Anthology, 2022. pp. 7171-7176 (COLING; 1).

@inproceedings{347e646761df4b7982a818bf19b17a12,

title = "Improving code-switched ASR with linguistic information",

abstract = "This paper seeks to improve the performance of automatic speech recognition (ASR) systems operating on code-switched speech. Code-switching refers to the alternation of languages within a conversation, a phenomenon that is of increasing importance considering the rapid rise in the number of bilingual speakers in the world. It is particularly challenging for ASR owing to the relative scarcity of code-switching speech and text data, even when the individual languages are themselves well-resourced. This paper proposes to overcome this challenge by applying linguistic theories in order to generate more realistic code-switching text, necessary for language modelling in ASR. Working with English-Spanish code-switching, we find that Equivalence Constraint theory and part-of-speech labelling are particularly helpful for text generation, and bring 2% improvement to ASR performance.",

author = "Jie Chi and Peter Bell",

year = "2022",

month = nov,

day = "3",

language = "English",

volume = "29",

series = "COLING",

publisher = "ACL Anthology",

number = "1",

pages = "7171--7176",

editor = "Nicoletta Calzolari and Chu-Ren Huang and Hansaem Kim and James Pustejovsky and Leo Wanner and Key-Sun Choi and Pum-Mo Ryu and Hsin-Hsi Chen and Lucia Donatelli and Heng Ji and Sadao Kurohashi and Patrizia Paggio and Nianwen Xue and Seokhwan Kim and Younggyun Hahm and Zhong He and Lee, {Tony Kyungil} and Enrico Santus and Francis Bond and Seung-Hoon Na",

booktitle = "Proceedings of the 29th International Conference on Computational Linguistics",

note = "The 29th International Conference on Computational Linguistics, 2022, COLING 2022 ; Conference date: 12-10-2022 Through 17-10-2022",

url = "https://coling2022.org/",

}

Chi, J & Bell, P 2022, Improving code-switched ASR with linguistic information. in N Calzolari, C-R Huang, H Kim, J Pustejovsky, L Wanner, K-S Choi, P-M Ryu, H-H Chen, L Donatelli, H Ji, S Kurohashi, P Paggio, N Xue, S Kim, Y Hahm, Z He, TK Lee, E Santus, F Bond & S-H Na (eds), Proceedings of the 29th International Conference on Computational Linguistics. vol. 29, COLING, no. 1, vol. 29, ACL Anthology, pp. 7171-7176, The 29th International Conference on Computational Linguistics, 2022, Gyeongju, Korea, Republic of, 12/10/22. <https://aclanthology.org/2022.coling-1.627/>

Improving code-switched ASR with linguistic information. / Chi, Jie; Bell, Peter.
Proceedings of the 29th International Conference on Computational Linguistics. ed. / Nicoletta Calzolari; Chu-Ren Huang; Hansaem Kim; James Pustejovsky; Leo Wanner; Key-Sun Choi; Pum-Mo Ryu; Hsin-Hsi Chen; Lucia Donatelli; Heng Ji; Sadao Kurohashi; Patrizia Paggio; Nianwen Xue; Seokhwan Kim; Younggyun Hahm; Zhong He; Tony Kyungil Lee; Enrico Santus; Francis Bond; Seung-Hoon Na. Vol. 29 ACL Anthology, 2022. p. 7171-7176 (COLING; Vol. 29, No. 1).

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

TY - GEN

T1 - Improving code-switched ASR with linguistic information

AU - Chi, Jie

AU - Bell, Peter

N1 - Conference code: 29

PY - 2022/11/3

Y1 - 2022/11/3

N2 - This paper seeks to improve the performance of automatic speech recognition (ASR) systems operating on code-switched speech. Code-switching refers to the alternation of languages within a conversation, a phenomenon that is of increasing importance considering the rapid rise in the number of bilingual speakers in the world. It is particularly challenging for ASR owing to the relative scarcity of code-switching speech and text data, even when the individual languages are themselves well-resourced. This paper proposes to overcome this challenge by applying linguistic theories in order to generate more realistic code-switching text, necessary for language modelling in ASR. Working with English-Spanish code-switching, we find that Equivalence Constraint theory and part-of-speech labelling are particularly helpful for text generation, and bring 2% improvement to ASR performance.

AB - This paper seeks to improve the performance of automatic speech recognition (ASR) systems operating on code-switched speech. Code-switching refers to the alternation of languages within a conversation, a phenomenon that is of increasing importance considering the rapid rise in the number of bilingual speakers in the world. It is particularly challenging for ASR owing to the relative scarcity of code-switching speech and text data, even when the individual languages are themselves well-resourced. This paper proposes to overcome this challenge by applying linguistic theories in order to generate more realistic code-switching text, necessary for language modelling in ASR. Working with English-Spanish code-switching, we find that Equivalence Constraint theory and part-of-speech labelling are particularly helpful for text generation, and bring 2% improvement to ASR performance.

M3 - Conference contribution

VL - 29

T3 - COLING

SP - 7171

EP - 7176

BT - Proceedings of the 29th International Conference on Computational Linguistics

A2 - Calzolari, Nicoletta

A2 - Huang, Chu-Ren

A2 - Kim, Hansaem

A2 - Pustejovsky, James

A2 - Wanner, Leo

A2 - Choi, Key-Sun

A2 - Ryu, Pum-Mo

A2 - Chen, Hsin-Hsi

A2 - Donatelli, Lucia

A2 - Ji, Heng

A2 - Kurohashi, Sadao

A2 - Paggio, Patrizia

A2 - Xue, Nianwen

A2 - Kim, Seokhwan

A2 - Hahm, Younggyun

A2 - He, Zhong

A2 - Lee, Tony Kyungil

A2 - Santus, Enrico

A2 - Bond, Francis

A2 - Na, Seung-Hoon

PB - ACL Anthology

T2 - The 29th International Conference on Computational Linguistics, 2022

Y2 - 12 October 2022 through 17 October 2022

ER -

Chi J, Bell P. Improving code-switched ASR with linguistic information. In Calzolari N, Huang CR, Kim H, Pustejovsky J, Wanner L, Choi KS, Ryu PM, Chen HH, Donatelli L, Ji H, Kurohashi S, Paggio P, Xue N, Kim S, Hahm Y, He Z, Lee TK, Santus E, Bond F, Na SH, editors, Proceedings of the 29th International Conference on Computational Linguistics. Vol. 29. ACL Anthology. 2022. p. 7171-7176. (COLING; 1).

Improving code-switched ASR with linguistic information

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this