Host: The Japanese Society for Artificial Intelligence
Name : The 33rd Annual Conference of the Japanese Society for Artificial Intelligence, 2019
Number : 33
Location : [in Japanese]
Date : June 04, 2019 - June 07, 2019
In this research, we regard a spoken dialogue systems as "dialogue media content" from user's view, and tries to realize dialogue media content circulation between creators and users. Focusing on the feedback from users to the creator of the content, an autonomous collection of dialogue profile and its effective feedback scheme is proposed. Experimental results showed that among the automatically collected feedback information, displaying recognition word list and occupation time at each state helps creators for content refinement and updates. Further interview with a professional creator after a long free-creation run also indicates that this sheme can motivate ones to create the content.
