2:00 PM - 2:20 PM
[1N3-GS-9-04] Motivational Interviewing Conversation Agents using LLM
Keywords:Agent, Dialogue Systems, Motivational Interviewing
The purpose of this research is to construct a counseling system for motivational interviewing. First, we created a ChatGPT-based utterance summarization function using three steps of prompts. As a dialogue engine, we fine-tuned an LLM to generate response as a counselor. By combining these two modules, we created a motivational interviewing dialogue system. Then, we integrated this dialogue system with ASR and TTS functions as speech input and output modules, and used a CG-character animation to produce embodied conversational agent. Finally, we discussed future directions to improve our system.
Authentication for paper PDF access
A password is required to view paper PDFs. If you are a registered participant, please log on the site from Participant Log In.
You could view the PDF with entering the PDF viewing password bellow.