Doing More with Less: Determining Optimal Pre-training Model for Irish Automatic Speech Recognition through Multi-step Fine-tuning

Caoilfhionn Ní Dheoráin, Ruth Holmes, Nicholas Evans, Thomas Laurent, Anthony Ventresque, Ellen Rushe


Abstract
In recent years, there has been an upsurge in research on automatic speech recognition (ASR) for low-resource languages. Particularly, transfer learning using multi-lingual models has become a popular remedy for the lack of available datasets for target languages. However, given the complexities associated with each individual language, we argue it is unlikely that a single multi-lingual pre-training model will provide equal performance gains across all languages. We also recognise the important, and insufficiently studied influence that the specific pre-training dataset has on the performance of the model. In this paper, using the Irish language as a case study, we propose a more directed, incremental form of pre-training which we term multi-step fine-tuning. This method accounts for the complex relationships between the language and dataset features of the source pre-training and target datasets. We show multi-step fine-tuning improves performance over simple multi-lingual fine-tuning alone, and we investigate factors leading to certain pre-trained models achieving better results through linguistic and dataset similarity measures. This research also investigates the uniformity of the performance gains across different demographics. We show that the optimal pre-training strategy can differ between demographics suggesting that more careful pre-training dataset selection is necessary to ensure equitable outcomes in practice.
Anthology ID:
2026.speakable-1.18
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:
162–173
Language:
External URL:
https://lrec.elra.info/lrec2026-ws-speakable-18
DOI:
10.63317/43cqo4vswrry
Bibkey:
Cite (ACL):
Caoilfhionn Ní Dheoráin, Ruth Holmes, Nicholas Evans, Thomas Laurent, Anthony Ventresque, and Ellen Rushe. 2026. Doing More with Less: Determining Optimal Pre-training Model for Irish Automatic Speech Recognition through Multi-step Fine-tuning. In Proceedings of Speech Language Models in Low-Resource Settings: Performance, Evaluation, and Bias Analysis (SPEAKABLE) @ LREC 2026, pages 162–173, Palma, Mallorca (Spain). ELRA Language Resources Association (ELRA).
Cite (Informal):
Doing More with Less: Determining Optimal Pre-training Model for Irish Automatic Speech Recognition through Multi-step Fine-tuning (Ní Dheoráin et al., SPEAKABLE 2026)
Copy Citation: