HOME / Lecture video / Research Talk
Research Talk

The coming ages when AI becomes our conversation partners
Cutting edge of conversational systems with large-scale neural network models
Hiroaki Sugiyama
Interaction Research Group, Innovative Communication Laboratory

Abstract

In recent years, social dialogue systems designed to satisfy people's desire for dialogue have been attracting attention. NTT has been researching various dialogue systems for a long time, including research on giving systems personalities and work on improving the smoothness of dialogue by organically linking multiple dialogue robots. On the other hand, with the recent rapid development of deep learning, English social dialog systems using large-scale deep learning have been proposed.
In this lecture, we introduce NTT's efforts to date, as well as details of NTT's latest deep learning-based Japanese social dialogue system and its achievements and remained issues.

[1] H. Sugiyama, T. Meguro, Y. Yoshikawa, J. Yamato, “Improving dialogue continuity using inter-robot interaction,” in Proc. 27th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN), pp. 105-112, 2018.
[2] H. Narimatsu, H. Sugiyama, M. Mizukami, T. Arimoto, “Rationale for Using Chat-Oriented Dialogue System’s Experience to Convey Empathy,” The Sixth Linguistic and Cognitive Approaches to Dialog Agents, 2021.
[3] H. Sugiyama, H. Narimatsu, M. Mizukami, T. Arimoto, Y. Chiba, T. Meguro, H. Nakajima, “Development of conversational system talking about hobby using Transformer-based encoder-decoder model,” SIG-SLUD, Vol. B5, No. 02, pp. 104-109, 2020 (in Japanese).

Speaker
Hiroaki Sugiyama
Hiroaki Sugiyama
Interaction Research Group, Innovative Communication Laboratory