Enhancing Standard and Dialectal Frisian ASR: Multilingual Fine-tuning and Language Identification for Improved Low-resource Performance
Збережено в:
| Опубліковано в:: | The Institute of Electrical and Electronics Engineers, Inc. (IEEE) Conference Proceedings (2025), p. 1-5 |
|---|---|
| Автор: | |
| Інші автори: | , , , , |
| Опубліковано: |
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
|
| Предмети: | |
| Онлайн доступ: | Citation/Abstract |
| Теги: |
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
MARC
| LEADER | 00000nab a2200000uu 4500 | ||
|---|---|---|---|
| 001 | 3268870229 | ||
| 003 | UK-CbPIL | ||
| 024 | 7 | |a 10.1109/ICASSP49660.2025.10889692 |2 doi | |
| 035 | |a 3268870229 | ||
| 045 | 2 | |b d20250101 |b d20251231 | |
| 084 | |a 228229 |2 nlm | ||
| 100 | 1 | |a Amooie, Reihaneh |u Center for Language and Cognition University of Groningen,Groningen,The Netherlands | |
| 245 | 1 | |a Enhancing Standard and Dialectal Frisian ASR: Multilingual Fine-tuning and Language Identification for Improved Low-resource Performance | |
| 260 | |b The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |c 2025 | ||
| 513 | |a Conference Proceedings | ||
| 520 | 3 | |a Conference Title: ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)Conference Start Date: 2025 April 6Conference End Date: 2025 April 11Conference Location: Hyderabad, IndiaAutomatic Speech Recognition (ASR) performance for low-resource languages is still far behind that of higher-resource languages such as English, due to a lack of sufficient labeled data. State-of-the-art methods deploy self-supervised transfer learning where a model pre-trained on large amounts of data is fine-tuned using little labeled data in a target low-resource language. In this paper, we present and examine a method for fine-tuning an SSL-based model in order to improve the performance for Frisian and its regional dialects (Clay Frisian, Wood Frisian, and South Frisian). We show that Frisian ASR performance can be improved by using multilingual (Frisian, Dutch, English and German) fine-tuning data and an auxiliary language identification task. In addition, our findings show that performance on dialectal speech suffers substantially, and, importantly, that this effect is moderated by the elicitation approach used to collect the dialectal data. Our findings also particularly suggest that relying solely on standard language data for ASR evaluation may underestimate real-world performance, particularly in languages with substantial dialectal variation. | |
| 653 | |a Performance enhancement | ||
| 653 | |a Multilingualism | ||
| 653 | |a Automatic speech recognition | ||
| 653 | |a English language | ||
| 653 | |a Languages | ||
| 653 | |a Acoustics | ||
| 653 | |a Economic | ||
| 700 | 1 | |a De Vries, Wietse |u Center for Language and Cognition University of Groningen,Groningen,The Netherlands | |
| 700 | 1 | |a Yun Hao |u Center for Language and Cognition University of Groningen,Groningen,The Netherlands | |
| 700 | 1 | |a Dijkstra, Jelske |u Mercator European Research Centre Fryske Akademy,Friesland,The Netherlands | |
| 700 | 1 | |a Coler, Matt |u Speech Technology Lab University of Groningen,Friesland,The Netherlands | |
| 700 | 1 | |a Wieling, Martijn |u Center for Language and Cognition University of Groningen,Groningen,The Netherlands | |
| 773 | 0 | |t The Institute of Electrical and Electronics Engineers, Inc. (IEEE) Conference Proceedings |g (2025), p. 1-5 | |
| 786 | 0 | |d ProQuest |t Science Database | |
| 856 | 4 | 1 | |3 Citation/Abstract |u https://www.proquest.com/docview/3268870229/abstract/embedded/7BTGNMKEMPT1V9Z2?source=fedsrch |