Can LLM Agents Identify Spoken Dialects like a Linguist?
2026
Read Can LLM Agents Identify Spoken Dialects like a Linguist? (2026) 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-2603-29541.txt.
Description
Bystrich and co-authors ask whether language model agents can identify spoken dialects like a linguist. Labeled dialect speech is scarce, which makes the task hard. Their test case is Swiss German. The agents work from phonetic transcripts made by speech recognition systems. They also get language resources like dialect feature maps and vowel history rules. With that linguistic help, the models' predictions improved. A human linguist baseline is included for comparison. The work shows transcripts can aid dialect classification.
Catalog record
| YEAR | 2026 |
|---|---|
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 29482 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.29541