Voice as a Digital Biomarker: Foundation Model-Based COPD Assessment

preprint OA: closed CC-BY-4.0
📄 Open PDF Full text JSON View at publisher
AI-generated deep summary by claude@2026-07, 2026-07-04 · read from full text

This preprint studied whether voice recordings can serve as a digital biomarker for diagnosing and staging chronic obstructive pulmonary disease (COPD) without spirometry, using a voice-only model based on self-supervised speech representations from wav2vec 2.0. The authors collected 1,709 recordings from 277 participants (227 with COPD and 50 controls) during maximum phonation and standardized reading tasks, obtained both before and after a 30-second chair-stand test, and trained/predicted COPD presence and severity without handcrafted acoustic features or clinical variables. The post-exercise reading condition performed best, with an AUC of 0.81 for COPD detection and 0.71 for severity classification; discrimination was stronger in adults under 65 years, while severity prediction was similar across age groups. As a preprint that is not peer reviewed, findings may be limited by the study design and reporting typical of early-stage work, and the paper does not explicitly discuss adenomyosis or endometriosis. The paper does not explicitly discuss endometriosis or adenomyosis; it was included in the corpus via a keyword match in the upstream search index.

Read from the paper's body, not the abstract. Not a substitute for reading the paper. No clinical advice. How this works

Abstract

Abstract Chronic obstructive pulmonary disease (COPD) is frequently underdiagnosed due to limited access to spirometry, highlighting the need for simple and scalable screening tools. Voice offers an easily obtainable and widely accessible respiratory biomarker, yet its potential for COPD assessment remains largely unexplored. Thus, we developed a voice-only COPD assessment model leveraging self-supervised speech representations from a wav2vec 2.0 foundation model using patient-recorded voice. We collected 1,709 recordings from 277 participants (227 COPD, 50 controls) across maximum phonation and standardized reading tasks captured both before and after a 30-second chair-stand test. Without handcrafted acoustic features or clinical variables, the model accurately predicted COPD presence and severity. The post-exercise reading condition achieved the highest performance, with an AUC of 0.81 for COPD detection and 0.71 for severity classification. Age-stratified analysis showed strong discrimination in adults younger than 65 years (AUC 0.87), while severity prediction remained consistent across age groups. These findings demonstrate that exertion-induced vocal alterations, combined with foundation-model representations, encode physiologic signatures of airflow limitation, enabling a practical and scalable approach for COPD screening and remote monitoring.
Full text 17,466 characters · extracted from preprint-html · click to expand
Voice as a Digital Biomarker: Foundation Model-Based COPD Assessment | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Voice as a Digital Biomarker: Foundation Model-Based COPD Assessment Sang Mee Lee, Hyein Ryu, Sunga Kong, Sun Hye Shin, Wooseong Huh, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8302274/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Chronic obstructive pulmonary disease (COPD) is frequently underdiagnosed due to limited access to spirometry, highlighting the need for simple and scalable screening tools. Voice offers an easily obtainable and widely accessible respiratory biomarker, yet its potential for COPD assessment remains largely unexplored. Thus, we developed a voice-only COPD assessment model leveraging self-supervised speech representations from a wav2vec 2.0 foundation model using patient-recorded voice. We collected 1,709 recordings from 277 participants (227 COPD, 50 controls) across maximum phonation and standardized reading tasks captured both before and after a 30-second chair-stand test. Without handcrafted acoustic features or clinical variables, the model accurately predicted COPD presence and severity. The post-exercise reading condition achieved the highest performance, with an AUC of 0.81 for COPD detection and 0.71 for severity classification. Age-stratified analysis showed strong discrimination in adults younger than 65 years (AUC 0.87), while severity prediction remained consistent across age groups. These findings demonstrate that exertion-induced vocal alterations, combined with foundation-model representations, encode physiologic signatures of airflow limitation, enabling a practical and scalable approach for COPD screening and remote monitoring. Health sciences/Biomarkers Health sciences/Diseases Health sciences/Health care Health sciences/Medical research Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8302274","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":561725777,"identity":"5181f656-d8c9-48cb-bc5d-ca9a8119447d","order_by":0,"name":"Sang Mee Lee","email":"","orcid":"","institution":"Sungkyunkwan University","correspondingAuthor":false,"prefix":"","firstName":"Sang","middleName":"Mee","lastName":"Lee","suffix":""},{"id":561725778,"identity":"a82ce1a2-3bbb-4dce-acc4-690d438a5b66","order_by":1,"name":"Hyein Ryu","email":"","orcid":"","institution":"Sungkyunkwan University","correspondingAuthor":false,"prefix":"","firstName":"Hyein","middleName":"","lastName":"Ryu","suffix":""},{"id":561725779,"identity":"83e5f02d-8a11-425c-b603-c5f5babbc767","order_by":2,"name":"Sunga Kong","email":"","orcid":"","institution":"Sungkyunkwan University","correspondingAuthor":false,"prefix":"","firstName":"Sunga","middleName":"","lastName":"Kong","suffix":""},{"id":561725780,"identity":"642ce361-d46f-4718-adba-5baec716a67f","order_by":3,"name":"Sun Hye Shin","email":"","orcid":"","institution":"Sungkyunkwan University","correspondingAuthor":false,"prefix":"","firstName":"Sun","middleName":"Hye","lastName":"Shin","suffix":""},{"id":561725781,"identity":"e1f122a9-5c5f-4785-8253-538143d41ff8","order_by":4,"name":"Wooseong Huh","email":"","orcid":"","institution":"Sungkyunkwan University","correspondingAuthor":false,"prefix":"","firstName":"Wooseong","middleName":"","lastName":"Huh","suffix":""},{"id":561725782,"identity":"934d66dc-3deb-4b91-b96c-6fcd7b48e798","order_by":5,"name":"Myung Jin Chung","email":"","orcid":"","institution":"Sungkyunkwan University","correspondingAuthor":false,"prefix":"","firstName":"Myung","middleName":"Jin","lastName":"Chung","suffix":""},{"id":561725783,"identity":"55f8c4e2-5257-4a40-b0fe-93ab64369a39","order_by":6,"name":"Juhee Cho","email":"","orcid":"","institution":"Sungkyunkwan University","correspondingAuthor":false,"prefix":"","firstName":"Juhee","middleName":"","lastName":"Cho","suffix":""},{"id":561725784,"identity":"6a6b0638-2815-4d35-9e4e-4fb64776c736","order_by":7,"name":"Taeyoung Kim","email":"","orcid":"","institution":"Samsung Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Taeyoung","middleName":"","lastName":"Kim","suffix":""},{"id":561725786,"identity":"691c126f-65ff-4811-aff3-55c5923dfccd","order_by":8,"name":"Hye Yun Park","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3UlEQVRIiWNgGAWjYBACAziLh/kAY8MBECuBaC1sCSRr4TEgTou5RPIxad4dd+TNe858/jjjzGEGfvYcA7xaLGekpUnznnlmOOds7zbJDTcOM0j2vMGvxeBGjpk0b9thxhn8vNsYH3w4DBIhpCX/G0iL/Qx+nscfQVrsCWvJYQNpSZzB28MAdpiBBCEtZ54ZW85te5Y8g+eYmeSMM+k8EmeeFeDXcjz54Y23bXdsZ/AkP/7Yc8xajr89eQNeLQwCCSwSDAwH4Hwe/MpBgP8A8wdkLaNgFIyCUTAKMAAA0qZQruKzjFkAAAAASUVORK5CYII=","orcid":"","institution":"Sungkyunkwan University","correspondingAuthor":true,"prefix":"","firstName":"Hye","middleName":"Yun","lastName":"Park","suffix":""}],"badges":[],"createdAt":"2025-12-07 23:53:16","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8302274/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8302274/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":98626945,"identity":"5cd3b6f2-9805-421a-adb9-77a2fc8373f2","added_by":"auto","created_at":"2025-12-19 17:10:03","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3408492,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript251206.docx","url":"https://assets-eu.researchsquare.com/files/rs-8302274/v1/b533c55388aba136b20720a3.docx"},{"id":98548528,"identity":"349d45d1-b6a7-466c-98c4-5697ad981b8f","added_by":"auto","created_at":"2025-12-18 19:49:09","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9968,"visible":true,"origin":"","legend":"","description":"","filename":"98aac80bf5e748a58147c2cbeda0ef6f.json","url":"https://assets-eu.researchsquare.com/files/rs-8302274/v1/518935d3899281cac461b0c6.json"},{"id":98548529,"identity":"d4c33b18-395c-4dee-9c08-9048d853c192","added_by":"auto","created_at":"2025-12-18 19:49:09","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":102329,"visible":true,"origin":"","legend":"","description":"","filename":"98aac80bf5e748a58147c2cbeda0ef6f1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-8302274/v1/9312ba92583df2707ddfeed7.xml"},{"id":98548534,"identity":"9b4129ff-5757-41a9-98a9-700bf950a0ca","added_by":"auto","created_at":"2025-12-18 19:49:09","extension":"png","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":161393,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8302274/v1/f4f06bb4e173a0964718947b.png"},{"id":98625440,"identity":"a6be9108-f28e-4146-af70-d5808453d380","added_by":"auto","created_at":"2025-12-19 17:09:07","extension":"png","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":107076,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8302274/v1/0bd661aed10b44acc34204c6.png"},{"id":98626107,"identity":"16aa001b-165f-46a6-8961-f5c0da11432b","added_by":"auto","created_at":"2025-12-19 17:09:32","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":139889,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8302274/v1/d83ae317e1d1f7094a2c9062.png"},{"id":98548537,"identity":"179f6f7d-431c-4b11-bf23-2caeffb57296","added_by":"auto","created_at":"2025-12-18 19:49:09","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":142344,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8302274/v1/acb8beebd9f7aa2227934d41.png"},{"id":98625919,"identity":"823509f8-91bf-48a8-9dd8-14444e17fa99","added_by":"auto","created_at":"2025-12-19 17:09:25","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":15454,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8302274/v1/564ba581a3b2b00d2a6a6589.png"},{"id":98548530,"identity":"2879c99c-6fbe-4df3-82ef-c549d770f2b7","added_by":"auto","created_at":"2025-12-18 19:49:09","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35248,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-8302274/v1/f8aab8862b3a19638d2c7386.png"},{"id":98548536,"identity":"ab837b06-bdfa-4da3-b03c-3cb50383bd87","added_by":"auto","created_at":"2025-12-18 19:49:09","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29457,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8302274/v1/50c299c897b16e83aa38205c.png"},{"id":98625998,"identity":"358cac2e-7d0e-46ff-9597-9ebae855c428","added_by":"auto","created_at":"2025-12-19 17:09:28","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29416,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-8302274/v1/1f4bd810ff969d6af73c5609.png"},{"id":98548540,"identity":"31a1d998-9bbc-4d7b-b59d-8fc85ea08d6e","added_by":"auto","created_at":"2025-12-18 19:49:09","extension":"xml","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":100404,"visible":true,"origin":"","legend":"","description":"","filename":"98aac80bf5e748a58147c2cbeda0ef6f1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-8302274/v1/fbb4f2ad017b2cbf2906dda6.xml"},{"id":98625437,"identity":"b6a08e78-3e82-486a-a279-9fd71ce09241","added_by":"auto","created_at":"2025-12-19 17:09:07","extension":"html","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":112512,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-8302274/v1/36e49e2e0894d430f7786fad.html"},{"id":103753048,"identity":"7f1c8aa2-9d4b-4995-b047-c7643dfd2fd7","added_by":"auto","created_at":"2026-03-02 13:27:30","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":735828,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript251206.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8302274/v1_covered_abca9359-f666-4642-a9cc-f3cbee8cb79c.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Voice as a Digital Biomarker: Foundation Model-Based COPD Assessment","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-8302274/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8302274/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eChronic obstructive pulmonary disease (COPD) is frequently underdiagnosed due to limited access to spirometry, highlighting the need for simple and scalable screening tools. Voice offers an easily obtainable and widely accessible respiratory biomarker, yet its potential for COPD assessment remains largely unexplored. Thus, we developed a voice-only COPD assessment model leveraging self-supervised speech representations from a wav2vec 2.0 foundation model using patient-recorded voice. We collected 1,709 recordings from 277 participants (227 COPD, 50 controls) across maximum phonation and standardized reading tasks captured both before and after a 30-second chair-stand test. Without handcrafted acoustic features or clinical variables, the model accurately predicted COPD presence and severity. The post-exercise reading condition achieved the highest performance, with an AUC of 0.81 for COPD detection and 0.71 for severity classification. Age-stratified analysis showed strong discrimination in adults younger than 65 years (AUC 0.87), while severity prediction remained consistent across age groups. These findings demonstrate that exertion-induced vocal alterations, combined with foundation-model representations, encode physiologic signatures of airflow limitation, enabling a practical and scalable approach for COPD screening and remote monitoring.\u003c/p\u003e","manuscriptTitle":"Voice as a Digital Biomarker: Foundation Model-Based COPD Assessment","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-12-18 19:49:04","doi":"10.21203/rs.3.rs-8302274/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"ab5d07d0-cbb0-44f1-ad2f-234faa3df2dd","owner":[],"postedDate":"December 18th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":59811395,"name":"Health sciences/Biomarkers"},{"id":59811396,"name":"Health sciences/Diseases"},{"id":59811397,"name":"Health sciences/Health care"},{"id":59811398,"name":"Health sciences/Medical research"}],"tags":[],"updatedAt":"2026-03-02T13:26:58+00:00","versionOfRecord":[],"versionCreatedAt":"2025-12-18 19:49:04","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8302274","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8302274","identity":"rs-8302274","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}

Text is read by the "Ask this paper" AI Q&A widget below. Extraction quality varies by source — PMC NXML preserves structure cleanly, OA-HTML may include some navigation residue, and OA-PDF can have broken hyphenation. The publisher copy (via DOI) is the canonical version.

My notes (saved in your browser only)

Ask this paper AI returns verbatim quotes from the full text · source: preprint-html

Answers must be backed by verbatim quotes from this paper's full text. Hallucinated quotes are dropped automatically; if no verbatim passage answers the question, we say so. How this works

Citation neighborhood (no data yet)

We don't have any in-corpus citations linked to this paper yet. This is a recent paper (2025) — citers typically take a year or two to land, and the OpenAlex reference graph may still be filling in.

Source provenance

europepmc
last seen: 2026-05-20T01:45:00.602351+00:00
unpaywall
last seen: 2026-05-22T02:00:06.705733+00:00
License: CC-BY-4.0