@inproceedings{koga-etal-2026-jphonabx,
title = "{JP}hon{ABX}: {J}apanese Phonological Discrimination Dataset for Moraic and Accentual Minimal Pairs",
author = "Koga, Yurie and
Kando, Shunsuke and
Miyao, Yusuke",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.1072/",
doi = "10.18653/2026.findings-emnlp.1072",
pages = "22676--22695",
ISBN = "979-8-89176-462-0",
abstract = "Self-supervised speech models are known to encode phonological information, yet it remains underexplored whether they capture phonological properties characteristic of Japanese, such as mora and pitch accent. This paper introduces JPhonABX, a Japanese phonological discrimination dataset targeting moraic and accentual contrasts. It consists of 192,000 moraic and 98,000 accentual ABX test items, covering various types of contrasts such as those involving moraic nasal, geminate consonant, vowel length, and pitch accent. Our human evaluation by native Japanese speakers shows high discrimination accuracy and inter-rater agreement, validating the quality of the dataset. We further evaluate several self-supervised speech models and find that while almost all of them show no significant difference from humans in mora discrimination, they perform significantly worse in accent discrimination, and even models trained on Japanese fail to reach human-level performance."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="koga-etal-2026-jphonabx">
<titleInfo>
<title>JPhonABX: Japanese Phonological Discrimination Dataset for Moraic and Accentual Minimal Pairs</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yurie</namePart>
<namePart type="family">Koga</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shunsuke</namePart>
<namePart type="family">Kando</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yusuke</namePart>
<namePart type="family">Miyao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Self-supervised speech models are known to encode phonological information, yet it remains underexplored whether they capture phonological properties characteristic of Japanese, such as mora and pitch accent. This paper introduces JPhonABX, a Japanese phonological discrimination dataset targeting moraic and accentual contrasts. It consists of 192,000 moraic and 98,000 accentual ABX test items, covering various types of contrasts such as those involving moraic nasal, geminate consonant, vowel length, and pitch accent. Our human evaluation by native Japanese speakers shows high discrimination accuracy and inter-rater agreement, validating the quality of the dataset. We further evaluate several self-supervised speech models and find that while almost all of them show no significant difference from humans in mora discrimination, they perform significantly worse in accent discrimination, and even models trained on Japanese fail to reach human-level performance.</abstract>
<identifier type="citekey">koga-etal-2026-jphonabx</identifier>
<identifier type="doi">10.18653/2026.findings-emnlp.1072</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.1072/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>22676</start>
<end>22695</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T JPhonABX: Japanese Phonological Discrimination Dataset for Moraic and Accentual Minimal Pairs
%A Koga, Yurie
%A Kando, Shunsuke
%A Miyao, Yusuke
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F koga-etal-2026-jphonabx
%X Self-supervised speech models are known to encode phonological information, yet it remains underexplored whether they capture phonological properties characteristic of Japanese, such as mora and pitch accent. This paper introduces JPhonABX, a Japanese phonological discrimination dataset targeting moraic and accentual contrasts. It consists of 192,000 moraic and 98,000 accentual ABX test items, covering various types of contrasts such as those involving moraic nasal, geminate consonant, vowel length, and pitch accent. Our human evaluation by native Japanese speakers shows high discrimination accuracy and inter-rater agreement, validating the quality of the dataset. We further evaluate several self-supervised speech models and find that while almost all of them show no significant difference from humans in mora discrimination, they perform significantly worse in accent discrimination, and even models trained on Japanese fail to reach human-level performance.
%R 10.18653/2026.findings-emnlp.1072
%U https://aclanthology.org/2026.findings-emnlp.1072/
%U https://doi.org/10.18653/2026.findings-emnlp.1072
%P 22676-22695
Markdown (Informal)
[JPhonABX: Japanese Phonological Discrimination Dataset for Moraic and Accentual Minimal Pairs](https://aclanthology.org/2026.findings-emnlp.1072/) (Koga et al., Findings 2026)
ACL