Adapting Foundational ASR Models to Efik: An Empirical Study of an Extremely Low-Resource Tonal Language

Offiong Bassey Edet, Stephen Orok Duke, Enoima Essien Umoh, Benjamin Okon Nyong, Andrew Asuquo Nkpanam


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.
Anthology ID:
2026.speakable-1.1
Volume:
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)
Editors:
Nina Hosseini-Kivanani, Alessio Brutti, Marco Matassoni, Sandipana Dowerah, Davide Liga, Christoph Schommer
Venues:
SPEAKABLE | WS
SIG:
Publisher:
ELRA Language Resources Association (ELRA)
Note:
Pages:
1–7
Language:
External URL:
https://lrec.elra.info/lrec2026-ws-speakable-01
DOI:
10.63317/3wmtoxexdady
Bibkey:
Cite (ACL):
Offiong Bassey Edet, Stephen Orok Duke, Enoima Essien Umoh, Benjamin Okon Nyong, and Andrew Asuquo Nkpanam. 2026. Adapting Foundational ASR Models to Efik: An Empirical Study of an Extremely Low-Resource Tonal Language. In Proceedings of Speech Language Models in Low-Resource Settings: Performance, Evaluation, and Bias Analysis (SPEAKABLE) @ LREC 2026, pages 1–7, Palma, Mallorca (Spain). ELRA Language Resources Association (ELRA).
Cite (Informal):
Adapting Foundational ASR Models to Efik: An Empirical Study of an Extremely Low-Resource Tonal Language (Edet et al., SPEAKABLE 2026)
Copy Citation: