Improving the Inclusivity of Dutch Speech Recognition by Fine-tuning Whisper on the JASMIN-CGN Corpus
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909508506222592 |
|---|---|
| author | Shekoufandeh, Golshid Boersma, Paul Bosch, Antal van den |
| author_facet | Shekoufandeh, Golshid Boersma, Paul Bosch, Antal van den |
| contents | We test and study the variation in speech recognition of fine-tuned versions of the Whisper model on child, elderly and non-native Dutch speech from the JASMIN-CGN corpus. Our primary goal is to evaluate how speakers' age and linguistic background influence Whisper's performance. Whisper achieves varying Word Error Rates (WER) when fine-tuned on subpopulations of specific ages and linguistic backgrounds. Fine-tuned performance is remarkably better than zero-shot performance, achieving a relative reduction in WER of 81% for native children, 72% for non-native children, 67% for non-native adults, and 65% for native elderly people. Our findings underscore the importance of training speech recognition models like Whisper on underrepresented subpopulations such as children, the elderly, and non-native speakers. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_17284 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Improving the Inclusivity of Dutch Speech Recognition by Fine-tuning Whisper on the JASMIN-CGN Corpus Shekoufandeh, Golshid Boersma, Paul Bosch, Antal van den Computation and Language Sound Audio and Speech Processing We test and study the variation in speech recognition of fine-tuned versions of the Whisper model on child, elderly and non-native Dutch speech from the JASMIN-CGN corpus. Our primary goal is to evaluate how speakers' age and linguistic background influence Whisper's performance. Whisper achieves varying Word Error Rates (WER) when fine-tuned on subpopulations of specific ages and linguistic backgrounds. Fine-tuned performance is remarkably better than zero-shot performance, achieving a relative reduction in WER of 81% for native children, 72% for non-native children, 67% for non-native adults, and 65% for native elderly people. Our findings underscore the importance of training speech recognition models like Whisper on underrepresented subpopulations such as children, the elderly, and non-native speakers. |
| title | Improving the Inclusivity of Dutch Speech Recognition by Fine-tuning Whisper on the JASMIN-CGN Corpus |
| topic | Computation and Language Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2502.17284 |