@inproceedings{hjortnaes-etal-2020-improving,
title = "Improving the Language Model for Low-Resource {ASR} with Online Text Corpora",
author = "Hjortnaes, Nils and
Arkhangelskiy, Timofey and
Partanen, Niko and
Rie{\ss}ler, Michael and
Tyers, Francis",
editor = "Beermann, Dorothee and
Besacier, Laurent and
Sakti, Sakriani and
Soria, Claudia",
booktitle = "Proceedings of the 1st Joint Workshop on Spoken Language Technologies for Under-resourced languages (SLTU) and Collaboration and Computing for Under-Resourced Languages (CCURL)",
month = may,
year = "2020",
address = "Marseille, France",
publisher = "European Language Resources association",
url = "https://aclanthology.org/2020.sltu-1.47/",
pages = "336--341",
language = "eng",
ISBN = "979-10-95546-35-1",
abstract = "In this paper, we expand on previous work on automatic speech recognition in a low-resource scenario typical of data collected by field linguists. We train DeepSpeech models on 35 hours of dialectal Komi speech recordings and correct the output using language models constructed from various sources. Previous experiments showed that transfer learning using DeepSpeech can improve the accuracy of a speech recognizer for Komi, though the error rate remained very high. In this paper we present further experiments with language models created using KenLM from text materials available online. These are constructed from two corpora, one containing literary texts, one for social media content, and another combining the two. We then trained the model using each language model to explore the impact of the language model data source on the speech recognition model. Our results show significant improvements of over 25{\%} in character error rate and nearly 20{\%} in word error rate. This offers important methodological insight into how ASR results can be improved under low-resource conditions: transfer learning can be used to compensate the lack of training data in the target language, and online texts are a very useful resource when developing language models in this context."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="hjortnaes-etal-2020-improving">
<titleInfo>
<title>Improving the Language Model for Low-Resource ASR with Online Text Corpora</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nils</namePart>
<namePart type="family">Hjortnaes</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Timofey</namePart>
<namePart type="family">Arkhangelskiy</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Niko</namePart>
<namePart type="family">Partanen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Michael</namePart>
<namePart type="family">Rießler</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Francis</namePart>
<namePart type="family">Tyers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<language>
<languageTerm type="text">eng</languageTerm>
</language>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 1st Joint Workshop on Spoken Language Technologies for Under-resourced languages (SLTU) and Collaboration and Computing for Under-Resourced Languages (CCURL)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dorothee</namePart>
<namePart type="family">Beermann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Laurent</namePart>
<namePart type="family">Besacier</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sakriani</namePart>
<namePart type="family">Sakti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Claudia</namePart>
<namePart type="family">Soria</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>European Language Resources association</publisher>
<place>
<placeTerm type="text">Marseille, France</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-10-95546-35-1</identifier>
</relatedItem>
<abstract>In this paper, we expand on previous work on automatic speech recognition in a low-resource scenario typical of data collected by field linguists. We train DeepSpeech models on 35 hours of dialectal Komi speech recordings and correct the output using language models constructed from various sources. Previous experiments showed that transfer learning using DeepSpeech can improve the accuracy of a speech recognizer for Komi, though the error rate remained very high. In this paper we present further experiments with language models created using KenLM from text materials available online. These are constructed from two corpora, one containing literary texts, one for social media content, and another combining the two. We then trained the model using each language model to explore the impact of the language model data source on the speech recognition model. Our results show significant improvements of over 25% in character error rate and nearly 20% in word error rate. This offers important methodological insight into how ASR results can be improved under low-resource conditions: transfer learning can be used to compensate the lack of training data in the target language, and online texts are a very useful resource when developing language models in this context.</abstract>
<identifier type="citekey">hjortnaes-etal-2020-improving</identifier>
<location>
<url>https://aclanthology.org/2020.sltu-1.47/</url>
</location>
<part>
<date>2020-05</date>
<extent unit="page">
<start>336</start>
<end>341</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Improving the Language Model for Low-Resource ASR with Online Text Corpora
%A Hjortnaes, Nils
%A Arkhangelskiy, Timofey
%A Partanen, Niko
%A Rießler, Michael
%A Tyers, Francis
%Y Beermann, Dorothee
%Y Besacier, Laurent
%Y Sakti, Sakriani
%Y Soria, Claudia
%S Proceedings of the 1st Joint Workshop on Spoken Language Technologies for Under-resourced languages (SLTU) and Collaboration and Computing for Under-Resourced Languages (CCURL)
%D 2020
%8 May
%I European Language Resources association
%C Marseille, France
%@ 979-10-95546-35-1
%G eng
%F hjortnaes-etal-2020-improving
%X In this paper, we expand on previous work on automatic speech recognition in a low-resource scenario typical of data collected by field linguists. We train DeepSpeech models on 35 hours of dialectal Komi speech recordings and correct the output using language models constructed from various sources. Previous experiments showed that transfer learning using DeepSpeech can improve the accuracy of a speech recognizer for Komi, though the error rate remained very high. In this paper we present further experiments with language models created using KenLM from text materials available online. These are constructed from two corpora, one containing literary texts, one for social media content, and another combining the two. We then trained the model using each language model to explore the impact of the language model data source on the speech recognition model. Our results show significant improvements of over 25% in character error rate and nearly 20% in word error rate. This offers important methodological insight into how ASR results can be improved under low-resource conditions: transfer learning can be used to compensate the lack of training data in the target language, and online texts are a very useful resource when developing language models in this context.
%U https://aclanthology.org/2020.sltu-1.47/
%P 336-341
Markdown (Informal)
[Improving the Language Model for Low-Resource ASR with Online Text Corpora](https://aclanthology.org/2020.sltu-1.47/) (Hjortnaes et al., SLTU 2020)
ACL
- Nils Hjortnaes, Timofey Arkhangelskiy, Niko Partanen, Michael Rießler, and Francis Tyers. 2020. Improving the Language Model for Low-Resource ASR with Online Text Corpora. In Proceedings of the 1st Joint Workshop on Spoken Language Technologies for Under-resourced languages (SLTU) and Collaboration and Computing for Under-Resourced Languages (CCURL), pages 336–341, Marseille, France. European Language Resources association.