@inproceedings{ogunremi-etal-2023-multilingual,
title = "Multilingual self-supervised speech representations improve the speech recognition of low-resource {A}frican languages with codeswitching",
author = "Ogunremi, Tolulope and
Manning, Christopher and
Jurafsky, Dan",
editor = "Winata, Genta and
Kar, Sudipta and
Zhukova, Marina and
Solorio, Thamar and
Diab, Mona and
Sitaram, Sunayana and
Choudhury, Monojit and
Bali, Kalika",
booktitle = "Proceedings of the 6th Workshop on Computational Approaches to Linguistic Code-Switching",
month = dec,
year = "2023",
address = "Singapore",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.calcs-1.8",
pages = "83--88",
abstract = "While many speakers of low-resource languages regularly code-switch between their languages and other regional languages or English, datasets of codeswitched speech are too small to train bespoke acoustic models from scratch or do language model rescoring. Here we propose finetuning self-supervised speech representations such as wav2vec 2.0 XLSR to recognize code-switched data. We find that finetuning self-supervised multilingual representations and augmenting them with n-gram language models trained from transcripts reduces absolute word error rates by up to 20{\%} compared to baselines of hybrid models trained from scratch on code-switched data. Our findings suggest that in circumstances with limited training data finetuning self-supervised representations is a better performing and viable solution.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ogunremi-etal-2023-multilingual">
<titleInfo>
<title>Multilingual self-supervised speech representations improve the speech recognition of low-resource African languages with codeswitching</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tolulope</namePart>
<namePart type="family">Ogunremi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christopher</namePart>
<namePart type="family">Manning</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dan</namePart>
<namePart type="family">Jurafsky</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 6th Workshop on Computational Approaches to Linguistic Code-Switching</title>
</titleInfo>
<name type="personal">
<namePart type="given">Genta</namePart>
<namePart type="family">Winata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sudipta</namePart>
<namePart type="family">Kar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marina</namePart>
<namePart type="family">Zhukova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Thamar</namePart>
<namePart type="family">Solorio</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mona</namePart>
<namePart type="family">Diab</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunayana</namePart>
<namePart type="family">Sitaram</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Monojit</namePart>
<namePart type="family">Choudhury</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kalika</namePart>
<namePart type="family">Bali</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Singapore</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>While many speakers of low-resource languages regularly code-switch between their languages and other regional languages or English, datasets of codeswitched speech are too small to train bespoke acoustic models from scratch or do language model rescoring. Here we propose finetuning self-supervised speech representations such as wav2vec 2.0 XLSR to recognize code-switched data. We find that finetuning self-supervised multilingual representations and augmenting them with n-gram language models trained from transcripts reduces absolute word error rates by up to 20% compared to baselines of hybrid models trained from scratch on code-switched data. Our findings suggest that in circumstances with limited training data finetuning self-supervised representations is a better performing and viable solution.</abstract>
<identifier type="citekey">ogunremi-etal-2023-multilingual</identifier>
<location>
<url>https://aclanthology.org/2023.calcs-1.8</url>
</location>
<part>
<date>2023-12</date>
<extent unit="page">
<start>83</start>
<end>88</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Multilingual self-supervised speech representations improve the speech recognition of low-resource African languages with codeswitching
%A Ogunremi, Tolulope
%A Manning, Christopher
%A Jurafsky, Dan
%Y Winata, Genta
%Y Kar, Sudipta
%Y Zhukova, Marina
%Y Solorio, Thamar
%Y Diab, Mona
%Y Sitaram, Sunayana
%Y Choudhury, Monojit
%Y Bali, Kalika
%S Proceedings of the 6th Workshop on Computational Approaches to Linguistic Code-Switching
%D 2023
%8 December
%I Association for Computational Linguistics
%C Singapore
%F ogunremi-etal-2023-multilingual
%X While many speakers of low-resource languages regularly code-switch between their languages and other regional languages or English, datasets of codeswitched speech are too small to train bespoke acoustic models from scratch or do language model rescoring. Here we propose finetuning self-supervised speech representations such as wav2vec 2.0 XLSR to recognize code-switched data. We find that finetuning self-supervised multilingual representations and augmenting them with n-gram language models trained from transcripts reduces absolute word error rates by up to 20% compared to baselines of hybrid models trained from scratch on code-switched data. Our findings suggest that in circumstances with limited training data finetuning self-supervised representations is a better performing and viable solution.
%U https://aclanthology.org/2023.calcs-1.8
%P 83-88
Markdown (Informal)
[Multilingual self-supervised speech representations improve the speech recognition of low-resource African languages with codeswitching](https://aclanthology.org/2023.calcs-1.8) (Ogunremi et al., CALCS 2023)
ACL