Voice-Based COVID-19 Detection: A Physiologically-Grounded Biomarker Approach and the Limits of Transferring the Widmark Alpha Constant - An Analysis of the Coswara-Data Respiratory Sound Corpus (IISc Bangalore)

{"Published":[0],"in":[1,25,153,203,212,220,265],"VoxMedica":[2],"Review":[3],"—":[4,37,52,269,302],"Inaugural":[5],"Special":[6],"Issue:":[7],"The":[8,136],"Vocal":[9],"Biomarker":[10],"Ecosystem":[11],"(Vol.":[12],"1,":[13,15],"No.":[14],"July":[16],"2026)":[17],"Abstract":[18],"This":[19],"study":[20,29],"adapts":[21],"the":[22,54,86,123,198,204,213,226,243,266,272,278,307],"methodology":[23],"developed":[24],"a":[26,45,100,157,167,233,256],"prior":[27],"voice-alcohol":[28],"(BAS":[30],"Alcohol":[31],"Language":[32],"Corpus,":[33],"\\"Promil":[34],"Avcısı\\"":[35],"v5)":[36],"physiologically":[38],"interpretable":[39],"acoustic":[40,299],"biomarkers,":[41],"leakage-free":[42],"evaluation,":[43],"and":[44,114,156,180,183,191,207,242,250,296],"sex/age-based":[46],"\\"alpha":[47],"constant\\"":[48],"(Widmark":[49],"r)":[50],"covariate":[51,147],"to":[53,253,259],"Coswara-Data":[55],"COVID-19":[56],"respiratory":[57],"sound":[58],"corpus":[59],"(Sharma":[60],"et":[61,65],"al.,":[62,66],"2020;":[63],"Bhattacharya":[64],"2023).":[67],"For":[68],"2,746":[69],"participants,":[70],"nine":[71],"recordings":[72],"each":[73],"(deep/shallow":[74],"breathing,":[75],"heavy/shallow":[76],"cough,":[77],"normal/fast":[78],"counting,":[79],"three":[80],"sustained":[81],"vowels)":[82],"were":[83,92],"extracted":[84],"from":[85,122],"full":[87],"34":[88],"GB":[89],"corpus;":[90],"biomarkers":[91,175,236,264],"computed":[93],"successfully":[94],"for":[95,162,188,195,231],"2,715":[96],"participants":[97,182,222],"(98.9%)":[98],"using":[99,298],"sound-type-specific":[101],"feature":[102],"set":[103],"(phonation:":[104],"F0,":[105],"jitter,":[106],"shimmer,":[107],"HNR,":[108],"formants;":[109],"speech:":[110],"rate/pause;":[111],"breathing:":[112],"envelope-periodicity":[113],"cycle":[115],"rate;":[116],"cough:":[117],"spectral/MFCC).":[118],"Critical":[119],"framing":[120],"difference":[121],"alcohol":[124,164,214,267,308],"study:":[125],"Coswara":[126],"contains":[127],"no":[128,149,210],"blood":[129,163],"or":[130],"serum":[131],"measurement":[132],"of":[133,235,246],"any":[134],"kind.":[135],"Widmark":[137],"r-factor":[138],"is":[139,216,229,248],"therefore":[140],"carried":[141],"over":[142],"as":[143,166],"an":[144],"explicitly":[145],"speculative":[146],"with":[148,209],"direct":[150],"pharmacokinetic":[151],"grounding":[152],"this":[154],"context,":[155],"self-reported":[158],"symptom-count":[159],"score":[160],"substitutes":[161],"concentration":[165],"continuous":[168],"\\"systemic":[169],"load\\"":[170],"proxy.":[171],"Findings:":[172],"(a)":[173],"numerous":[174],"differ":[176],"significantly":[177],"between":[178],"healthy":[179],"COVID-positive":[181,221],"remain":[184],"significant":[185,230],"after":[186],"adjusting":[187],"age,":[189],"sex,":[190],"smoking":[192],"status":[193],"(p<0.001":[194],"most);":[196],"(b)":[197],"single":[199],"largest":[200],"effect":[201,245],"size":[202],"entire":[205],"study,":[206],"one":[208],"counterpart":[211],"work,":[215],"reduced":[217],"breathing-envelope":[218],"periodicity":[219],"(Cohen's":[223],"d=-0.565);":[224],"(c)":[225],"symptom-load×r":[227],"interaction":[228],"only":[232],"handful":[234],"(F0,":[237],"formant":[238],"dispersion,":[239],"speech":[240],"rate)":[241],"correlation-strengthening":[244],"r-normalization":[247],"small":[249],"inconsistent":[251],"(+0.00":[252],"+0.03,":[254],"versus":[255],"consistent":[257],"+0.03":[258],"+0.06":[260],"across":[261],"nearly":[262],"all":[263],"study)":[268],"indicating":[270],"that":[271],"alpha":[273],"constant":[274],"does":[275],"not":[276],"carry":[277],"same":[279],"explanatory":[280],"weight":[281],"here;":[282],"(d)":[283],"binary":[284],"classification":[285],"(healthy":[286],"vs.":[287],"COVID-positive)":[288],"reaches":[289],"AUC=0.856±0.021":[290],"(Gradient":[291],"Boosting)":[292],"including":[293],"demographic":[294],"covariates,":[295],"AUC=0.853±0.023":[297],"features":[300],"alone":[301],"both":[303],"substantially":[304],"stronger":[305],"than":[306],"study's":[309],"AUC=0.710.":[310]}

Authors

Institutions

Publication Details

Journal
Open MIND
Published
2026-07-15
DOI
https://doi.org/10.5281/zenodo.21322921
Primary Topic
Respiratory and Cough-Related Research
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
article

Voice-Based COVID-19 Detection: A Physiologically-Grounded Biomarker Approach and the Limits of Transferring the Widmark Alpha Constant - An Analysis of the Coswara-Data Respiratory Sound Corpus (IISc Bangalore)

Gökhan Aydemir, Oğuzhan AYDEMİR
Open MIND
Respiratory and Cough-Related Research
article

Voice-Based COVID-19 Detection: A Physiologically-Grounded Biomarker Approach and the Limits of Transferring the Widmark Alpha Constant - An Analysis of the Coswara-Data Respiratory Sound Corpus (IISc Bangalore)

Gökhan Aydemir, Oğuzhan AYDEMİR
article en

Abstract

Published in VoxMedica Review — Inaugural Special Issue: The Vocal Biomarker Ecosystem (Vol. 1, No. 1, July 2026) Abstract This study adapts the methodology developed in a prior voice-alcohol study (BAS Alcohol Language Corpus, "Promil Avcısı" v5) — physiologically interpretable acoustic biomarkers, leakage-free evaluation, and a sex/age-based "alpha constant" (Widmark r) covariate — to the Coswara-Data COVID-19 respiratory sound corpus (Sharma et al., 2020; Bhattacharya et al., 2023). For 2,746 participants, nine recordings each (deep/shallow breathing, heavy/shallow cough, normal/fast counting, three sustained vowels) were extracted from the full 34 GB corpus; biomarkers were computed successfully for 2,715 participants (98.9%) using a sound-type-specific feature set (phonation: F0, jitter, shimmer, HNR, formants; speech: rate/pause; breathing: envelope-periodicity and cycle rate; cough: spectral/MFCC). Critical framing difference from the alcohol study: Coswara contains no blood or serum measurement of any kind. The Widmark r-factor is therefore carried over as an explicitly speculative covariate with no direct pharmacokinetic grounding in this context, and a self-reported symptom-count score substitutes for blood alcohol concentration as a continuous "systemic load" proxy. Findings: (a) numerous biomarkers differ significantly between healthy and COVID-positive participants and remain significant after adjusting for age, sex, and smoking status (p<0.001 for most); (b) the single largest effect size in the entire study, and one with no counterpart in the alcohol work, is reduced breathing-envelope periodicity in COVID-positive participants (Cohen's d=-0.565); (c) the symptom-load×r interaction is significant for only a handful of biomarkers (F0, formant dispersion, speech rate) and the correlation-strengthening effect of r-normalization is small and inconsistent (+0.00 to +0.03, versus a consistent +0.03 to +0.06 across nearly all biomarkers in the alcohol study) — indicating that the alpha constant does not carry the same explanatory weight here; (d) binary classification (healthy vs. COVID-positive) reaches AUC=0.856±0.021 (Gradient Boosting) including demographic covariates, and AUC=0.853±0.023 using acoustic features alone — both substantially stronger than the alcohol study's AUC=0.710.

Open MIND
Cox & Company (United States) (US)
Openalex Percentile: Top 9%
Respiratory and Cough-Related Research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.