Author_Institution :
Dept. of Inf. & Comput. Eng., Kanazawa Inst. of Technol., Nonoichi, Japan
Abstract :
Speech dialogue systems, such as Apple´s “Siri,” have gradually become more widespread, and in the near future, a greater number of general users will have the opportunity to communicate with such systems. To facilitate this, it is necessary for the system to communicate more naturally with users, and to realize that both verbal and nonverbal information must be taken into consideration. Therefore, a model that can select contextually appropriate nonverbal information should be developed. However, the relation between context and nonverbal information has not been sufficiently analyzed, because it is difficult to control the context of communication in an experiment. In this research, we clarify the relation between nonverbal information, such as the utterance rhythm, and the context of a dialogue by analyzing the correlation of utterance durations with the game state in a kind of “prisoner´s dilemma” that we introduce to control the context. The results show diverse correlations across four game states (Cooperation-Cooperation, Cooperation-Betrayal, Betrayal-Cooperation, Betrayal-Betrayal); however, in the Cooperation-Cooperation state, a positive correlation is often observed between the duration of utterances. From these results, we discuss the mechanism of synchrony between utterance rhythms.
Keywords :
human computer interaction; speech-based user interfaces; contextually appropriate nonverbal information; cooperation game; cooperation-cooperation state; game states; repetitive dialogue; speech dialogue systems; utterance duration; utterance rhythm synchrony; verbal information; Context; Correlation; Games; Humans; Speech; Switches; Time series analysis;