@inproceedings{kang-etal-2026-higraph,
title = "{H}i{G}raph: Explicit Discrepancy Reasoning with Hierarchical Graph Propagation for Joint Pronunciation Assessment",
author = "Kang, Hyunku and
Choi, Eunseo and
Kim, Ireh and
Kim, Chanwoo",
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.1060/",
pages = "22441--22455",
ISBN = "979-8-89176-462-0",
abstract = "Automated pronunciation assessment predicts pronunciation quality at the phone, word, and utterance levels. Recent joint models use both canonical and realized phone representations, but their mismatch is often encoded implicitly rather than preserved as a separate scoring path. This makes it difficult to trace how local phone errors affect word- and utterance-level scores. We propose HiGraph, which explicitly computes canonical-realized mismatch and propagates it through a hierarchical phone-to-word-to-utterance graph. This creates a structured discrepancy path from phone-level mismatch evidence to higher-level score representations. Under the same pipeline as our reproduced HMamba baseline, HiGraph improves word stress and utterance completeness on SpeechOcean762 (SO762). The learned word-level trace score also aligns with mispronunciation detection and diagnosis (MDD) ground truth. When words are ranked by this trace score, the top-ranked word contains a true mispronunciation in 81.5{\%} of test utterances, and at least one of the top two words does so in 93.8{\%} of test utterances. These results show that explicit discrepancy propagation can improve selected assessment dimensions while supporting feedback localization grounded in annotated mispronunciations."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kang-etal-2026-higraph">
<titleInfo>
<title>HiGraph: Explicit Discrepancy Reasoning with Hierarchical Graph Propagation for Joint Pronunciation Assessment</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hyunku</namePart>
<namePart type="family">Kang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Eunseo</namePart>
<namePart type="family">Choi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ireh</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chanwoo</namePart>
<namePart type="family">Kim</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>Automated pronunciation assessment predicts pronunciation quality at the phone, word, and utterance levels. Recent joint models use both canonical and realized phone representations, but their mismatch is often encoded implicitly rather than preserved as a separate scoring path. This makes it difficult to trace how local phone errors affect word- and utterance-level scores. We propose HiGraph, which explicitly computes canonical-realized mismatch and propagates it through a hierarchical phone-to-word-to-utterance graph. This creates a structured discrepancy path from phone-level mismatch evidence to higher-level score representations. Under the same pipeline as our reproduced HMamba baseline, HiGraph improves word stress and utterance completeness on SpeechOcean762 (SO762). The learned word-level trace score also aligns with mispronunciation detection and diagnosis (MDD) ground truth. When words are ranked by this trace score, the top-ranked word contains a true mispronunciation in 81.5% of test utterances, and at least one of the top two words does so in 93.8% of test utterances. These results show that explicit discrepancy propagation can improve selected assessment dimensions while supporting feedback localization grounded in annotated mispronunciations.</abstract>
<identifier type="citekey">kang-etal-2026-higraph</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.1060/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>22441</start>
<end>22455</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T HiGraph: Explicit Discrepancy Reasoning with Hierarchical Graph Propagation for Joint Pronunciation Assessment
%A Kang, Hyunku
%A Choi, Eunseo
%A Kim, Ireh
%A Kim, Chanwoo
%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 kang-etal-2026-higraph
%X Automated pronunciation assessment predicts pronunciation quality at the phone, word, and utterance levels. Recent joint models use both canonical and realized phone representations, but their mismatch is often encoded implicitly rather than preserved as a separate scoring path. This makes it difficult to trace how local phone errors affect word- and utterance-level scores. We propose HiGraph, which explicitly computes canonical-realized mismatch and propagates it through a hierarchical phone-to-word-to-utterance graph. This creates a structured discrepancy path from phone-level mismatch evidence to higher-level score representations. Under the same pipeline as our reproduced HMamba baseline, HiGraph improves word stress and utterance completeness on SpeechOcean762 (SO762). The learned word-level trace score also aligns with mispronunciation detection and diagnosis (MDD) ground truth. When words are ranked by this trace score, the top-ranked word contains a true mispronunciation in 81.5% of test utterances, and at least one of the top two words does so in 93.8% of test utterances. These results show that explicit discrepancy propagation can improve selected assessment dimensions while supporting feedback localization grounded in annotated mispronunciations.
%U https://aclanthology.org/2026.findings-emnlp.1060/
%P 22441-22455
Markdown (Informal)
[HiGraph: Explicit Discrepancy Reasoning with Hierarchical Graph Propagation for Joint Pronunciation Assessment](https://aclanthology.org/2026.findings-emnlp.1060/) (Kang et al., Findings 2026)
ACL