Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.07038 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912966848282624 |
|---|---|
| author | Rafkin, Emma DeGenaro, Dan Yang, Xiulin |
| author_facet | Rafkin, Emma DeGenaro, Dan Yang, Xiulin |
| contents | The development of resource-constrained approaches to automatic speech recognition (ASR) is of great interest due to its broad applicability to many low-resource languages for which there is scant usable data. Existing approaches to many low-resource natural language processing tasks leverage additional data from higher-resource languages that are closely related to a target low-resource language. One increasingly popular approach uses task arithmetic to combine models trained on different tasks to create a model for a task where there is little to no training data. In this paper, we consider training on a particular language to be a task, and we generate task vectors by fine-tuning variants of the Whisper ASR system. For pairs of high- and low-resource languages, we merge task vectors via a linear combination which is optimized on the downstream word error rate on the low-resource target language's validation set. Across 23 low-resource target languages for which we evaluate this technique, we find consistent word error rate improvements of up to 10% compared to a baseline without our approach. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_07038 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Task Arithmetic with Support Languages for Low-Resource ASR Rafkin, Emma DeGenaro, Dan Yang, Xiulin Computation and Language The development of resource-constrained approaches to automatic speech recognition (ASR) is of great interest due to its broad applicability to many low-resource languages for which there is scant usable data. Existing approaches to many low-resource natural language processing tasks leverage additional data from higher-resource languages that are closely related to a target low-resource language. One increasingly popular approach uses task arithmetic to combine models trained on different tasks to create a model for a task where there is little to no training data. In this paper, we consider training on a particular language to be a task, and we generate task vectors by fine-tuning variants of the Whisper ASR system. For pairs of high- and low-resource languages, we merge task vectors via a linear combination which is optimized on the downstream word error rate on the low-resource target language's validation set. Across 23 low-resource target languages for which we evaluate this technique, we find consistent word error rate improvements of up to 10% compared to a baseline without our approach. |
| title | Task Arithmetic with Support Languages for Low-Resource ASR |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2601.07038 |