@inproceedings{lee-etal-2025-dypcl,
title = "{D}y{PCL}: Dynamic Phoneme-level Contrastive Learning for Dysarthric Speech Recognition",
author = "Lee, Wonjun and
Im, Solee and
Do, Heejin and
Kim, Yunsu and
Ok, Jungseul and
Lee, Gary",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.naacl-long.240/",
doi = "10.18653/v1/2025.naacl-long.240",
pages = "4701--4712",
ISBN = "979-8-89176-189-6",
abstract = "Dysarthric speech recognition often suffers from performance degradation due to the intrinsic diversity of dysarthric severity and extrinsic disparity from normal speech. To bridge these gaps, we propose a Dynamic Phoneme-level Contrastive Learning (DyPCL) method, which leads to obtaining invariant representations across diverse speakers. We decompose the speech utterance into phoneme segments for phoneme-level contrastive learning, leveraging dynamic connectionist temporal classification alignment. Unlike prior studies focusing on utterance-level embeddings, our granular learning allows discrimination of subtle parts of speech. In addition, we introduce dynamic curriculum learning, which progressively transitions from easy negative samples to difficult-to-distinguishable negative samples based on phonetic similarity of phoneme. Our approach to training by difficulty levels alleviates the inherent variability of speakers, better identifying challenging speeches. Evaluated on the UASpeech dataset, DyPCL outperforms baseline models, achieving an average 22.10{\%} relative reduction in word error rate (WER) across the overall dysarthria group."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lee-etal-2025-dypcl">
<titleInfo>
<title>DyPCL: Dynamic Phoneme-level Contrastive Learning for Dysarthric Speech Recognition</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wonjun</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Solee</namePart>
<namePart type="family">Im</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Heejin</namePart>
<namePart type="family">Do</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yunsu</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jungseul</namePart>
<namePart type="family">Ok</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gary</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luis</namePart>
<namePart type="family">Chiruzzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-189-6</identifier>
</relatedItem>
<abstract>Dysarthric speech recognition often suffers from performance degradation due to the intrinsic diversity of dysarthric severity and extrinsic disparity from normal speech. To bridge these gaps, we propose a Dynamic Phoneme-level Contrastive Learning (DyPCL) method, which leads to obtaining invariant representations across diverse speakers. We decompose the speech utterance into phoneme segments for phoneme-level contrastive learning, leveraging dynamic connectionist temporal classification alignment. Unlike prior studies focusing on utterance-level embeddings, our granular learning allows discrimination of subtle parts of speech. In addition, we introduce dynamic curriculum learning, which progressively transitions from easy negative samples to difficult-to-distinguishable negative samples based on phonetic similarity of phoneme. Our approach to training by difficulty levels alleviates the inherent variability of speakers, better identifying challenging speeches. Evaluated on the UASpeech dataset, DyPCL outperforms baseline models, achieving an average 22.10% relative reduction in word error rate (WER) across the overall dysarthria group.</abstract>
<identifier type="citekey">lee-etal-2025-dypcl</identifier>
<identifier type="doi">10.18653/v1/2025.naacl-long.240</identifier>
<location>
<url>https://aclanthology.org/2025.naacl-long.240/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>4701</start>
<end>4712</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T DyPCL: Dynamic Phoneme-level Contrastive Learning for Dysarthric Speech Recognition
%A Lee, Wonjun
%A Im, Solee
%A Do, Heejin
%A Kim, Yunsu
%A Ok, Jungseul
%A Lee, Gary
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-189-6
%F lee-etal-2025-dypcl
%X Dysarthric speech recognition often suffers from performance degradation due to the intrinsic diversity of dysarthric severity and extrinsic disparity from normal speech. To bridge these gaps, we propose a Dynamic Phoneme-level Contrastive Learning (DyPCL) method, which leads to obtaining invariant representations across diverse speakers. We decompose the speech utterance into phoneme segments for phoneme-level contrastive learning, leveraging dynamic connectionist temporal classification alignment. Unlike prior studies focusing on utterance-level embeddings, our granular learning allows discrimination of subtle parts of speech. In addition, we introduce dynamic curriculum learning, which progressively transitions from easy negative samples to difficult-to-distinguishable negative samples based on phonetic similarity of phoneme. Our approach to training by difficulty levels alleviates the inherent variability of speakers, better identifying challenging speeches. Evaluated on the UASpeech dataset, DyPCL outperforms baseline models, achieving an average 22.10% relative reduction in word error rate (WER) across the overall dysarthria group.
%R 10.18653/v1/2025.naacl-long.240
%U https://aclanthology.org/2025.naacl-long.240/
%U https://doi.org/10.18653/v1/2025.naacl-long.240
%P 4701-4712
Markdown (Informal)
[DyPCL: Dynamic Phoneme-level Contrastive Learning for Dysarthric Speech Recognition](https://aclanthology.org/2025.naacl-long.240/) (Lee et al., NAACL 2025)
ACL
- Wonjun Lee, Solee Im, Heejin Do, Yunsu Kim, Jungseul Ok, and Gary Lee. 2025. DyPCL: Dynamic Phoneme-level Contrastive Learning for Dysarthric Speech Recognition. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 4701–4712, Albuquerque, New Mexico. Association for Computational Linguistics.