<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="kim-etal-2026-auditing">
    <titleInfo>
        <title>Auditing Speaker-Level Metadata Utility in Text-Residual Prosody Prediction</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Gaeun</namePart>
        <namePart type="family">Kim</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Jihyun</namePart>
        <namePart type="family">Park</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Sohee</namePart>
        <namePart type="family">Jang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Joon-Hyuk</namePart>
        <namePart type="family">Chang</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>Text only partially determines prosodic realization, leaving residual variation that may align with speaker-level metadata such as regional accent or first-language (L1) background. However, because these labels are assigned to speakers, their predictive gains may overlap with broader speaker-associated structure rather than uniquely reflect category-level prosodic effects. We frame this as an attribution problem and propose a speaker-controlled audit using matched text-residual prosody predictors that differ only in access to metadata labels and speaker identity. The audit measures label utility beyond text, speaker utility beyond text, and the additional empirical utility of the label when speaker conditioning is already available. Across VCTK regional-accent labels and L2-ARCTIC L1 labels, metadata yields modest gains over text, while speaker conditioning yields larger gains and leaves little additional label utility under speaker control. Random speaker-group and label-perturbation controls further characterize the specificity and use of the supplied metadata. Importantly, speaker-held-out evaluation shows that metadata transfer to unseen speakers is limited and target-dependent, indicating that observed-speaker metadata gains do not by themselves establish speaker-independent category-level effects. These results position speaker-controlled auditing as a prerequisite attribution check before assigning category-level interpretations to metadata-conditioned predictive gains.</abstract>
    <identifier type="citekey">kim-etal-2026-auditing</identifier>
    <location>
        <url>https://aclanthology.org/2026.findings-emnlp.2212/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>47393</start>
            <end>47412</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Auditing Speaker-Level Metadata Utility in Text-Residual Prosody Prediction
%A Kim, Gaeun
%A Park, Jihyun
%A Jang, Sohee
%A Chang, Joon-Hyuk
%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 kim-etal-2026-auditing
%X Text only partially determines prosodic realization, leaving residual variation that may align with speaker-level metadata such as regional accent or first-language (L1) background. However, because these labels are assigned to speakers, their predictive gains may overlap with broader speaker-associated structure rather than uniquely reflect category-level prosodic effects. We frame this as an attribution problem and propose a speaker-controlled audit using matched text-residual prosody predictors that differ only in access to metadata labels and speaker identity. The audit measures label utility beyond text, speaker utility beyond text, and the additional empirical utility of the label when speaker conditioning is already available. Across VCTK regional-accent labels and L2-ARCTIC L1 labels, metadata yields modest gains over text, while speaker conditioning yields larger gains and leaves little additional label utility under speaker control. Random speaker-group and label-perturbation controls further characterize the specificity and use of the supplied metadata. Importantly, speaker-held-out evaluation shows that metadata transfer to unseen speakers is limited and target-dependent, indicating that observed-speaker metadata gains do not by themselves establish speaker-independent category-level effects. These results position speaker-controlled auditing as a prerequisite attribution check before assigning category-level interpretations to metadata-conditioned predictive gains.
%U https://aclanthology.org/2026.findings-emnlp.2212/
%P 47393-47412

Markdown (Informal)

[Auditing Speaker-Level Metadata Utility in Text-Residual Prosody Prediction](https://aclanthology.org/2026.findings-emnlp.2212/) (Kim et al., Findings 2026)

• Auditing Speaker-Level Metadata Utility in Text-Residual Prosody Prediction (Kim et al., Findings 2026)

ACL

• Gaeun Kim, Jihyun Park, Sohee Jang, and Joon-Hyuk Chang. 2026. Auditing Speaker-Level Metadata Utility in Text-Residual Prosody Prediction. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 47393–47412, Budapest, Hungary. Association for Computational Linguistics.