Effects of Time Lag of Utterances to Communicative Actions on Embodied Interaction With Robot and CG Character

Abstract

The timing involved in generating communicative actions and utterances in a face-to-face greeting interaction for application in robot–human and computer-generated (CG) character–human interaction support systems is analyzed by synthesis. First, an analysis of human greeting clarifies the average pause and the average time delay in the utterance to a communicative action. Then, a synthesis-based analysis is performed by using an embodied robot system. This analysis confirms that the variation in the pause and the lag in the utterance to communicative actions produce different communicative effects, for example, a lag of about 0.3 sec is desirable for a familiar greeting and a longer lag is appropriate for a polite greeting. In addition, the synthesis-based analysis performed on a CG character system confirms the timing control effects. These results demonstrate the importance of timing control in embodied interactions as well as the applicability of such interactions in advanced communications with robots and CG characters.