<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bar-etal-2025-swiss">
    <titleInfo>
        <title>Swiss German Speech Translation and the Curse of Multidialectality</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Martin</namePart>
        <namePart type="family">Bär</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Andrea</namePart>
        <namePart type="family">DeMarco</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Gorka</namePart>
        <namePart type="family">Labaka</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2025-07</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the 22nd International Conference on Spoken Language Translation (IWSLT 2025)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Elizabeth</namePart>
            <namePart type="family">Salesky</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Marcello</namePart>
            <namePart type="family">Federico</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Antonis</namePart>
            <namePart type="family">Anastasopoulos</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>Association for Computational Linguistics</publisher>
            <place>
                <placeTerm type="text">Vienna, Austria (in-person and online)</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
        <identifier type="isbn">979-8-89176-272-5</identifier>
    </relatedItem>
    <abstract>In many languages, non-standardized varieties make the development of NLP models challenging. This paper explores various fine-tuning techniques and data setups for training Swiss German to Standard German speech-to-text translation models. While fine-tuning on all available Swiss German data yields the best results, ASR pre-training lowers performance by 1.48 BLEU points, and jointly training on Swiss and Standard German data reduces it by 2.29 BLEU. Our dialect transfer experiments suggest that an equivalent of the Curse of Multilinguality (Conneau et al., 2020) exists in dialectal speech processing, as training on multiple dialects jointly tends to decrease single-dialect performance. However, introducing small amounts of dialectal variability can improve the performance for low-resource dialects.</abstract>
    <identifier type="citekey">bar-etal-2025-swiss</identifier>
    <identifier type="doi">10.18653/v1/2025.iwslt-1.15</identifier>
    <location>
        <url>https://aclanthology.org/2025.iwslt-1.15/</url>
    </location>
    <part>
        <date>2025-07</date>
        <extent unit="page">
            <start>165</start>
            <end>179</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T Swiss German Speech Translation and the Curse of Multidialectality
%A Bär, Martin
%A DeMarco, Andrea
%A Labaka, Gorka
%Y Salesky, Elizabeth
%Y Federico, Marcello
%Y Anastasopoulos, Antonis
%S Proceedings of the 22nd International Conference on Spoken Language Translation (IWSLT 2025)
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria (in-person and online)
%@ 979-8-89176-272-5
%F bar-etal-2025-swiss
%X In many languages, non-standardized varieties make the development of NLP models challenging. This paper explores various fine-tuning techniques and data setups for training Swiss German to Standard German speech-to-text translation models. While fine-tuning on all available Swiss German data yields the best results, ASR pre-training lowers performance by 1.48 BLEU points, and jointly training on Swiss and Standard German data reduces it by 2.29 BLEU. Our dialect transfer experiments suggest that an equivalent of the Curse of Multilinguality (Conneau et al., 2020) exists in dialectal speech processing, as training on multiple dialects jointly tends to decrease single-dialect performance. However, introducing small amounts of dialectal variability can improve the performance for low-resource dialects.
%R 10.18653/v1/2025.iwslt-1.15
%U https://aclanthology.org/2025.iwslt-1.15/
%U https://doi.org/10.18653/v1/2025.iwslt-1.15
%P 165-179

Markdown (Informal)

[Swiss German Speech Translation and the Curse of Multidialectality](https://aclanthology.org/2025.iwslt-1.15/) (Bär et al., IWSLT 2025)

• Swiss German Speech Translation and the Curse of Multidialectality (Bär et al., IWSLT 2025)

ACL

• Martin Bär, Andrea DeMarco, and Gorka Labaka. 2025. Swiss German Speech Translation and the Curse of Multidialectality. In Proceedings of the 22nd International Conference on Spoken Language Translation (IWSLT 2025), pages 165–179, Vienna, Austria (in-person and online). Association for Computational Linguistics.