Application of Large Language Models to Multilevel Analysis of Narrative Speech
Martina Altamura, Francesco Petriglia, Giuseppe Mancini, Andrea Marini
In conclusion, despite the limitations of the current pilot study, LLMs appear to be a highly promising tool for the future. However, at present, the level of agreement measured does not allow for their use without careful supervision by an experienced clinician.
Keywords
Artificial intelligence, ChatGPT, Multilevel linguistic analysis, Narrative speech, Language assessment.
