@inproceedings{edet-etal-2026-adapting,
title = "Adapting Foundational {ASR} Models to {E}fik: An Empirical Study of an Extremely Low-Resource Tonal Language",
author = "Edet, Offiong Bassey and
Duke, Stephen Orok and
Umoh, Enoima Essien and
Nyong, Benjamin Okon and
Nkpanam, Andrew Asuquo",
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-1.1/",
doi = "10.63317/3wmtoxexdady",
pages = "1--7",
abstract = "Automatic Speech Recognition (ASR) has significantly transformed human-computer-interaction and natural language processing. However, many African spoken languages, including Efik, remain severely underrepresented in ASR research. This paper investigates the adoption of state-of-the-art foundational ASR models such as XLS-R and Whisper through fine-tuning for Efik, a low-resource tonal language and empirically evaluates their performance. We curate a 3-hour Efik speech dataset and conduct a comparative evaluation using standard ASR metrics. We further augmented the XLS-R CTC model with a 3-gram KenLM language model trained on an Efik text corpus. Experimental results show that XLS-R-300M + KenLM achieves a word error rate (WER) of 10.86{\%} and a character error rate (CER) of 3.16{\%}, substantially outperforming both the baseline XLS-R (WER: 29.2{\%}, CER: 6.4{\%}) and Whisper across noisy and multi-speaker conditions. These findings suggest that lightweight CTC models augmented with language model integration offer a more robust and practical approach for extremely low-resource tonal languages than larger sequence-to-sequence models."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="edet-etal-2026-adapting">
<titleInfo>
<title>Adapting Foundational ASR Models to Efik: An Empirical Study of an Extremely Low-Resource Tonal Language</title>
</titleInfo>
<name type="personal">
<namePart type="given">Offiong</namePart>
<namePart type="given">Bassey</namePart>
<namePart type="family">Edet</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Stephen</namePart>
<namePart type="given">Orok</namePart>
<namePart type="family">Duke</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Enoima</namePart>
<namePart type="given">Essien</namePart>
<namePart type="family">Umoh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Benjamin</namePart>
<namePart type="given">Okon</namePart>
<namePart type="family">Nyong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andrew</namePart>
<namePart type="given">Asuquo</namePart>
<namePart type="family">Nkpanam</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) has significantly transformed human-computer-interaction and natural language processing. However, many African spoken languages, including Efik, remain severely underrepresented in ASR research. This paper investigates the adoption of state-of-the-art foundational ASR models such as XLS-R and Whisper through fine-tuning for Efik, a low-resource tonal language and empirically evaluates their performance. We curate a 3-hour Efik speech dataset and conduct a comparative evaluation using standard ASR metrics. We further augmented the XLS-R CTC model with a 3-gram KenLM language model trained on an Efik text corpus. Experimental results show that XLS-R-300M + KenLM achieves a word error rate (WER) of 10.86% and a character error rate (CER) of 3.16%, substantially outperforming both the baseline XLS-R (WER: 29.2%, CER: 6.4%) and Whisper across noisy and multi-speaker conditions. These findings suggest that lightweight CTC models augmented with language model integration offer a more robust and practical approach for extremely low-resource tonal languages than larger sequence-to-sequence models.</abstract>
<identifier type="citekey">edet-etal-2026-adapting</identifier>
<identifier type="doi">10.63317/3wmtoxexdady</identifier>
<location>
<url>https://aclanthology.org/2026.speakable-1.1/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>1</start>
<end>7</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Adapting Foundational ASR Models to Efik: An Empirical Study of an Extremely Low-Resource Tonal Language
%A Edet, Offiong Bassey
%A Duke, Stephen Orok
%A Umoh, Enoima Essien
%A Nyong, Benjamin Okon
%A Nkpanam, Andrew Asuquo
%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 edet-etal-2026-adapting
%X Automatic Speech Recognition (ASR) has significantly transformed human-computer-interaction and natural language processing. However, many African spoken languages, including Efik, remain severely underrepresented in ASR research. This paper investigates the adoption of state-of-the-art foundational ASR models such as XLS-R and Whisper through fine-tuning for Efik, a low-resource tonal language and empirically evaluates their performance. We curate a 3-hour Efik speech dataset and conduct a comparative evaluation using standard ASR metrics. We further augmented the XLS-R CTC model with a 3-gram KenLM language model trained on an Efik text corpus. Experimental results show that XLS-R-300M + KenLM achieves a word error rate (WER) of 10.86% and a character error rate (CER) of 3.16%, substantially outperforming both the baseline XLS-R (WER: 29.2%, CER: 6.4%) and Whisper across noisy and multi-speaker conditions. These findings suggest that lightweight CTC models augmented with language model integration offer a more robust and practical approach for extremely low-resource tonal languages than larger sequence-to-sequence models.
%R 10.63317/3wmtoxexdady
%U https://aclanthology.org/2026.speakable-1.1/
%U https://doi.org/10.63317/3wmtoxexdady
%P 1-7
Markdown (Informal)
[Adapting Foundational ASR Models to Efik: An Empirical Study of an Extremely Low-Resource Tonal Language](https://aclanthology.org/2026.speakable-1.1/) (Edet et al., SPEAKABLE 2026)
ACL