<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bonilla-2026-beyond">
    <titleInfo>
        <title>Beyond Acoustics: Isolating Dialectal and Sociolinguistic Bias in Spanish ASR</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Johnatan</namePart>
        <namePart type="given">E</namePart>
        <namePart type="family">Bonilla</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the Seventh Workshop on Natural Language Processing and Computational Social Science</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Dallas</namePart>
            <namePart type="family">Card</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Anjalie</namePart>
            <namePart type="family">Field</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Katherine</namePart>
            <namePart type="family">Keith</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Julia</namePart>
            <namePart type="family">Mendelsohn</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">San Diego</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-426-2</identifier>
    </relatedItem>
    <abstract>Large-scale ASR systems such as Whisper achieve competitive aggregate Word Error Rate (WER) on multilingual benchmarks, but this aggregate conceals systematic disparities across speaker populations. We evaluate Whisper large-v3 on 276 recordings from the Corpus Oral y Sonoro del Español Rural (COSER), a dialectological archive of elderly rural speakers across all Spanish provinces. WER is computed separately for Informants and Interviewers within each recording, revealing that mixed-role evaluation underestimates Informant WER in the majority of provinces, with the largest corrections in southern areas.Negative Binomial regression with cluster-robust standar errors shows that Andalusia and Extremadura generate significantly more Informant errors than the Castilian heartland (Andalusia IRR = 1.20, p &lt; 0.001; Extremadura IRR = 1.24, p = 0.020), while no geographic predictor reaches significance for Interviewers sharing the same recording environment. Male Informants generate 12.5% more errors than females after geographic adjustment (p &lt; 0.001), consistent with differential vernacular retention in traditional rural communities. The geographic pattern aligns with established dialectological classifications of Peninsular Spanish. These results demonstrate that role-disaggregated evaluation is a necessary methodological prerequisite for fairness audits of ASR systems applied to sociolinguistically diverse corpora: aggregate benchmarks systematically suppress disparities that are borne disproportionately by the most underrepresented speaker populations, and their use in isolation constitutes both an allocative harm and a measurement failure</abstract>
    <identifier type="citekey">bonilla-2026-beyond</identifier>
    <location>
        <url>https://aclanthology.org/2026.nlpcss-1.8/</url>
    </location>
    <part>
        <date>2026-07</date>
        <extent unit="page">
            <start>123</start>
            <end>132</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Beyond Acoustics: Isolating Dialectal and Sociolinguistic Bias in Spanish ASR
%A Bonilla, Johnatan E.
%Y Card, Dallas
%Y Field, Anjalie
%Y Keith, Katherine
%Y Mendelsohn, Julia
%S Proceedings of the Seventh Workshop on Natural Language Processing and Computational Social Science
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego
%@ 979-8-89176-426-2
%F bonilla-2026-beyond
%X Large-scale ASR systems such as Whisper achieve competitive aggregate Word Error Rate (WER) on multilingual benchmarks, but this aggregate conceals systematic disparities across speaker populations. We evaluate Whisper large-v3 on 276 recordings from the Corpus Oral y Sonoro del Español Rural (COSER), a dialectological archive of elderly rural speakers across all Spanish provinces. WER is computed separately for Informants and Interviewers within each recording, revealing that mixed-role evaluation underestimates Informant WER in the majority of provinces, with the largest corrections in southern areas.Negative Binomial regression with cluster-robust standar errors shows that Andalusia and Extremadura generate significantly more Informant errors than the Castilian heartland (Andalusia IRR = 1.20, p < 0.001; Extremadura IRR = 1.24, p = 0.020), while no geographic predictor reaches significance for Interviewers sharing the same recording environment. Male Informants generate 12.5% more errors than females after geographic adjustment (p < 0.001), consistent with differential vernacular retention in traditional rural communities. The geographic pattern aligns with established dialectological classifications of Peninsular Spanish. These results demonstrate that role-disaggregated evaluation is a necessary methodological prerequisite for fairness audits of ASR systems applied to sociolinguistically diverse corpora: aggregate benchmarks systematically suppress disparities that are borne disproportionately by the most underrepresented speaker populations, and their use in isolation constitutes both an allocative harm and a measurement failure
%U https://aclanthology.org/2026.nlpcss-1.8/
%P 123-132

Markdown (Informal)

[Beyond Acoustics: Isolating Dialectal and Sociolinguistic Bias in Spanish ASR](https://aclanthology.org/2026.nlpcss-1.8/) (Bonilla, NLP+CSS 2026)

• Beyond Acoustics: Isolating Dialectal and Sociolinguistic Bias in Spanish ASR (Bonilla, NLP+CSS 2026)

ACL

• Johnatan E. Bonilla. 2026. Beyond Acoustics: Isolating Dialectal and Sociolinguistic Bias in Spanish ASR. In Proceedings of the Seventh Workshop on Natural Language Processing and Computational Social Science, pages 123–132, San Diego. Association for Computational Linguistics.