Adapting Whisper for Regional Dialects: Enhancing Public Services for Vulnerable Populations in the United Kingdom
Melissa Torgbi*, Andrew Clayman*, Jordan J. Speight, Harish Tayyar Madabushi, [0.3cm] Department of Computer Science, University of Bath, UK · 2025
Read Adapting Whisper for Regional Dialects: Enhancing Public Services for Vulnerable Populations in the United Kingdom by Melissa Torgbi*, Andrew Clayman*, Jordan J. Speight, Harish Tayyar Madabushi, [0.3cm] Department of Computer Science, University of Bath, UK (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-08502.txt.
Description
Melissa Torgbi and co-authors tested speech software on Scottish accents for public services in this 2025 study. Staff who take calls must hear callers clearly, yet speech tools can fail on local speech and harm those who most need help. The team gathered new speech in a service setting from two areas of Scotland.
A leading open speech model made more word errors on the new data than on standard test data. Extra training on speech from one area cut errors there, and it also helped on the other area, which suggests some gains carry across areas. Hand checks of errors showed that a single word error score hides both gains and new flaws, so close reading still matters.
Catalog record
| CREATOR | Melissa Torgbi*, Andrew Clayman*, Jordan J. Speight, Harish Tayyar Madabushi, [0.3cm] Department of Computer Science, University of Bath, UK |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-08502