Can LLM Agents Identify Spoken Dialects like a Linguist?
This work addresses dialect identification for linguists and language researchers, but it is incremental as it builds on existing methods with added linguistic features.
The paper tackled the challenge of audio dialect classification, particularly for Swiss German, by exploring whether LLM agents with linguistic resources can match HuBERT's performance, finding that LLM predictions improve with linguistic information and that automatic transcriptions aid classification but need refinement.
Due to the scarcity of labeled dialectal speech, audio dialect classification is a challenging task for most languages, including Swiss German. In this work, we explore the ability of large language models (LLMs) as agents in understanding the dialects and whether they can show comparable performance to models such as HuBERT in dialect classification. In addition, we provide an LLM baseline and a human linguist one. Our approach uses phonetic transcriptions produced by ASR systems and combines them with linguistic resources such as dialect feature maps, vowel history, and rules. Our findings indicate that, when linguistic information is provided, the LLM predictions improve. The human baseline shows that automatically generated transcriptions can be beneficial for such classifications, but also presents opportunities for improvement.