@inproceedings{lag-etal-2026-stage,
title = "Stage-Aware Cross-Lingual Transfer for {F}aroese {ASR}: When and Which Languages Matter",
author = "L{\'a}g, D{\'a}vid {\'i} and
Scalvini, Barbara and
Mena, Carlos Daniel and
Gu{\dh}nason, J{\'o}n",
editor = "Hosseini-Kivanani, Nina and
Brutti, Alessio and
Matassoni, Marco and
Dowerah, Sandipana and
Liga, Davide and
Schommer, Christoph",
booktitle = "Proceedings of Speech Language Models in Low-Resource Settings: Performance, Evaluation, and Bias Analysis ({SPEAKABLE}) @ {LREC} 2026",
month = may,
year = "2026",
address = "Palma, Mallorca (Spain)",
publisher = "ELRA Language Resources Association (ELRA)",
url = "https://aclanthology.org/2026.speakable-1.17/",
doi = "10.63317/5hpzgtgaxf43",
pages = "150--161",
abstract = "Automatic speech recognition (ASR) for low-resource languages remains challenging due to limited labeled data. Although multilingual models and the inclusion of related auxiliary languages enable cross-lingual transfer, it is still unclear how introducing cross-lingual information at different training stages-pre-training versus fine-tuning-affects downstream performance. Prior work largely treats transfer as a single-stage optimization problem without disentangling stage effects. We present a stage-aware analysis of cross-lingual transfer for Faroese ASR using related auxiliary languages and Wav2Vec 2.0 XLS-R models. We systematically compare two complementary adaptation pipelines: (i) cross-lingual supervised fine-tuning and (ii) cross-lingual continuous pre-training prior to fine-tuning. Both strategies are evaluated under a unified setup with controlled model architectures, balanced representation of auxiliary languages, and identical evaluation protocols. Results demonstrate that cross-lingual transfer is stage-dependent. Supervised adaptation optimizes in-domain accuracy, while pretraining-level adaptation enhances robustness and reduces Character Error Rate (CER). Auxiliary language effects vary across pipelines, reinforcing the idea that transfer effectiveness depends on when and how cross-lingual information is introduced. Comparisons with large-scale multilingual ASR models highlight trade-offs between model scale and explicit, small-scale domain-aware adaptation. These findings suggest that effective cross-lingual transfer for Faroese low-resource ASR is inherently stage-dependent rather than a single-step design choice."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lag-etal-2026-stage">
<titleInfo>
<title>Stage-Aware Cross-Lingual Transfer for Faroese ASR: When and Which Languages Matter</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dávid</namePart>
<namePart type="given">í</namePart>
<namePart type="family">Lág</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Barbara</namePart>
<namePart type="family">Scalvini</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carlos</namePart>
<namePart type="given">Daniel</namePart>
<namePart type="family">Mena</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jón</namePart>
<namePart type="family">Gu\dhnason</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of Speech Language Models in Low-Resource Settings: Performance, Evaluation, and Bias Analysis (SPEAKABLE) @ LREC 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nina</namePart>
<namePart type="family">Hosseini-Kivanani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alessio</namePart>
<namePart type="family">Brutti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marco</namePart>
<namePart type="family">Matassoni</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sandipana</namePart>
<namePart type="family">Dowerah</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Davide</namePart>
<namePart type="family">Liga</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christoph</namePart>
<namePart type="family">Schommer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resources Association (ELRA)</publisher>
<place>
<placeTerm type="text">Palma, Mallorca (Spain)</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Automatic speech recognition (ASR) for low-resource languages remains challenging due to limited labeled data. Although multilingual models and the inclusion of related auxiliary languages enable cross-lingual transfer, it is still unclear how introducing cross-lingual information at different training stages-pre-training versus fine-tuning-affects downstream performance. Prior work largely treats transfer as a single-stage optimization problem without disentangling stage effects. We present a stage-aware analysis of cross-lingual transfer for Faroese ASR using related auxiliary languages and Wav2Vec 2.0 XLS-R models. We systematically compare two complementary adaptation pipelines: (i) cross-lingual supervised fine-tuning and (ii) cross-lingual continuous pre-training prior to fine-tuning. Both strategies are evaluated under a unified setup with controlled model architectures, balanced representation of auxiliary languages, and identical evaluation protocols. Results demonstrate that cross-lingual transfer is stage-dependent. Supervised adaptation optimizes in-domain accuracy, while pretraining-level adaptation enhances robustness and reduces Character Error Rate (CER). Auxiliary language effects vary across pipelines, reinforcing the idea that transfer effectiveness depends on when and how cross-lingual information is introduced. Comparisons with large-scale multilingual ASR models highlight trade-offs between model scale and explicit, small-scale domain-aware adaptation. These findings suggest that effective cross-lingual transfer for Faroese low-resource ASR is inherently stage-dependent rather than a single-step design choice.</abstract>
<identifier type="citekey">lag-etal-2026-stage</identifier>
<identifier type="doi">10.63317/5hpzgtgaxf43</identifier>
<location>
<url>https://aclanthology.org/2026.speakable-1.17/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>150</start>
<end>161</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Stage-Aware Cross-Lingual Transfer for Faroese ASR: When and Which Languages Matter
%A Lág, Dávid í
%A Scalvini, Barbara
%A Mena, Carlos Daniel
%A Gu\dhnason, Jón
%Y Hosseini-Kivanani, Nina
%Y Brutti, Alessio
%Y Matassoni, Marco
%Y Dowerah, Sandipana
%Y Liga, Davide
%Y Schommer, Christoph
%S Proceedings of Speech Language Models in Low-Resource Settings: Performance, Evaluation, and Bias Analysis (SPEAKABLE) @ LREC 2026
%D 2026
%8 May
%I ELRA Language Resources Association (ELRA)
%C Palma, Mallorca (Spain)
%F lag-etal-2026-stage
%X Automatic speech recognition (ASR) for low-resource languages remains challenging due to limited labeled data. Although multilingual models and the inclusion of related auxiliary languages enable cross-lingual transfer, it is still unclear how introducing cross-lingual information at different training stages-pre-training versus fine-tuning-affects downstream performance. Prior work largely treats transfer as a single-stage optimization problem without disentangling stage effects. We present a stage-aware analysis of cross-lingual transfer for Faroese ASR using related auxiliary languages and Wav2Vec 2.0 XLS-R models. We systematically compare two complementary adaptation pipelines: (i) cross-lingual supervised fine-tuning and (ii) cross-lingual continuous pre-training prior to fine-tuning. Both strategies are evaluated under a unified setup with controlled model architectures, balanced representation of auxiliary languages, and identical evaluation protocols. Results demonstrate that cross-lingual transfer is stage-dependent. Supervised adaptation optimizes in-domain accuracy, while pretraining-level adaptation enhances robustness and reduces Character Error Rate (CER). Auxiliary language effects vary across pipelines, reinforcing the idea that transfer effectiveness depends on when and how cross-lingual information is introduced. Comparisons with large-scale multilingual ASR models highlight trade-offs between model scale and explicit, small-scale domain-aware adaptation. These findings suggest that effective cross-lingual transfer for Faroese low-resource ASR is inherently stage-dependent rather than a single-step design choice.
%R 10.63317/5hpzgtgaxf43
%U https://aclanthology.org/2026.speakable-1.17/
%U https://doi.org/10.63317/5hpzgtgaxf43
%P 150-161
Markdown (Informal)
[Stage-Aware Cross-Lingual Transfer for Faroese ASR: When and Which Languages Matter](https://aclanthology.org/2026.speakable-1.17/) (Lág et al., SPEAKABLE 2026)
ACL