CONFERENCE (INTERNATIONAL) An Open-Domain Avatar Chatbot by Exploiting a Large Language Model

Takato Yamazaki, Tomoya Mizumoto, Katsumasa Yoshikawa, Masaya Ohagi, Toshiki Kawamoto (LINE/Tokyo Institute of Technology), Toshinori Sato

24th Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL 2023)

September 11, 2023

With the ambition to create avatars capable of human-level casual conversation, we developed an open-domain avatar chatbot, situated in a virtual reality environment, that employs a large language model (LLM). Introducing the LLM posed several challenges for multimodal integration, such as developing techniques to align diverse outputs and avatar control, as well as addressing the issue of slow generation speed. To address these challenges, we integrated various external modules into our system. Our system is based on the award-winning model from the Dialogue System Live Competition 5. Through this work, we hope to stimulate discussions within the research community about the potential and challenges of multimodal dialogue systems enhanced with LLMs.

Paper : An Open-Domain Avatar Chatbot by Exploiting a Large Language Modelopen into new tab or window (external link)