@inproceedings{matsuura-etal-2026-adapting,
title = "Adapting Pretrained Models to Endangered Languages in {J}apan: A Comparative Study on Ryukyuan and {A}inu Speech Recognition",
author = "Matsuura, Kohei and
Ashihara, Takanori and
Kawahara, Tatsuya",
editor = "Piperidis, Stelios and
Bel, N{\'u}ria and
van den Heuvel, Henk and
Ide, Nancy and
Krek, Simon and
Toral, Antonio",
booktitle = "Proceedings of the Fifteenth Language Resources and Evaluation Conference",
month = may,
year = "2026",
address = "Palma de Mallorca, Spain",
publisher = "ELRA Language Resource Association",
url = "https://aclanthology.org/2026.lrec-1.112/",
doi = "10.63317/3iw5dymnwwbr",
pages = "1455--1463",
abstract = "We investigate high-accuracy and speaker-robust automatic speech recognition (ASR) models by leveraging pretrained models for endangered languages in Japan {---} Ryukyuan (Shuri dialect) and Ainu (Saru dialect) {---} to support language and cultural preservation. In particular, this study presents the first experimental study on building and evaluating an ASR model for the Ryukyuan language. Specifically, we compare existing multilingual pretrained models, Whisper and XLS-R, with our in-house Japanese-focused model (JP-90k) pretrained solely on a large-scale weakly-supervised Japanese dataset. These models were fine-tuned on up to 10 and 32 hours of Ryukyuan and Ainu data, respectively. As a result, JP-90k consistently outperformed other models of the similar size in both languages. In addition, it demonstrated a remarkable advantage when training data was very limited, i.e., an hour or less. These findings suggest that large-scale pretraining on a language closely related to the target ones can yield robust low-resource ASR, including for unseen speakers and out-of-domain conditions. Furthermore, we found that all pretrained models achieved convergence in ASR accuracy with as little as 3-5 hours of fine-tuning data for both languages."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="matsuura-etal-2026-adapting">
<titleInfo>
<title>Adapting Pretrained Models to Endangered Languages in Japan: A Comparative Study on Ryukyuan and Ainu Speech Recognition</title>
</titleInfo>
<name type="personal">
<namePart type="given">Kohei</namePart>
<namePart type="family">Matsuura</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Takanori</namePart>
<namePart type="family">Ashihara</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tatsuya</namePart>
<namePart type="family">Kawahara</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fifteenth Language Resources and Evaluation Conference</title>
</titleInfo>
<name type="personal">
<namePart type="given">Stelios</namePart>
<namePart type="family">Piperidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Núria</namePart>
<namePart type="family">Bel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Henk</namePart>
<namePart type="family">van den Heuvel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nancy</namePart>
<namePart type="family">Ide</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Simon</namePart>
<namePart type="family">Krek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antonio</namePart>
<namePart type="family">Toral</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resource Association</publisher>
<place>
<placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We investigate high-accuracy and speaker-robust automatic speech recognition (ASR) models by leveraging pretrained models for endangered languages in Japan — Ryukyuan (Shuri dialect) and Ainu (Saru dialect) — to support language and cultural preservation. In particular, this study presents the first experimental study on building and evaluating an ASR model for the Ryukyuan language. Specifically, we compare existing multilingual pretrained models, Whisper and XLS-R, with our in-house Japanese-focused model (JP-90k) pretrained solely on a large-scale weakly-supervised Japanese dataset. These models were fine-tuned on up to 10 and 32 hours of Ryukyuan and Ainu data, respectively. As a result, JP-90k consistently outperformed other models of the similar size in both languages. In addition, it demonstrated a remarkable advantage when training data was very limited, i.e., an hour or less. These findings suggest that large-scale pretraining on a language closely related to the target ones can yield robust low-resource ASR, including for unseen speakers and out-of-domain conditions. Furthermore, we found that all pretrained models achieved convergence in ASR accuracy with as little as 3-5 hours of fine-tuning data for both languages.</abstract>
<identifier type="citekey">matsuura-etal-2026-adapting</identifier>
<identifier type="doi">10.63317/3iw5dymnwwbr</identifier>
<location>
<url>https://aclanthology.org/2026.lrec-1.112/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>1455</start>
<end>1463</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Adapting Pretrained Models to Endangered Languages in Japan: A Comparative Study on Ryukyuan and Ainu Speech Recognition
%A Matsuura, Kohei
%A Ashihara, Takanori
%A Kawahara, Tatsuya
%Y Piperidis, Stelios
%Y Bel, Núria
%Y van den Heuvel, Henk
%Y Ide, Nancy
%Y Krek, Simon
%Y Toral, Antonio
%S Proceedings of the Fifteenth Language Resources and Evaluation Conference
%D 2026
%8 May
%I ELRA Language Resource Association
%C Palma de Mallorca, Spain
%F matsuura-etal-2026-adapting
%X We investigate high-accuracy and speaker-robust automatic speech recognition (ASR) models by leveraging pretrained models for endangered languages in Japan — Ryukyuan (Shuri dialect) and Ainu (Saru dialect) — to support language and cultural preservation. In particular, this study presents the first experimental study on building and evaluating an ASR model for the Ryukyuan language. Specifically, we compare existing multilingual pretrained models, Whisper and XLS-R, with our in-house Japanese-focused model (JP-90k) pretrained solely on a large-scale weakly-supervised Japanese dataset. These models were fine-tuned on up to 10 and 32 hours of Ryukyuan and Ainu data, respectively. As a result, JP-90k consistently outperformed other models of the similar size in both languages. In addition, it demonstrated a remarkable advantage when training data was very limited, i.e., an hour or less. These findings suggest that large-scale pretraining on a language closely related to the target ones can yield robust low-resource ASR, including for unseen speakers and out-of-domain conditions. Furthermore, we found that all pretrained models achieved convergence in ASR accuracy with as little as 3-5 hours of fine-tuning data for both languages.
%R 10.63317/3iw5dymnwwbr
%U https://aclanthology.org/2026.lrec-1.112/
%U https://doi.org/10.63317/3iw5dymnwwbr
%P 1455-1463
Markdown (Informal)
[Adapting Pretrained Models to Endangered Languages in Japan: A Comparative Study on Ryukyuan and Ainu Speech Recognition](https://aclanthology.org/2026.lrec-1.112/) (Matsuura et al., LREC 2026)
ACL