This paper describes the results of our research on a prototype of the conversation support system capable of activating group conversation in a vehicle. The goal of this system was to enable further activating a group conversation. Based on methodology used in existing technology of utterance analysis, we estimated the intentions and desires of each group member in a vehicle from their conversation, aiming to enhance the overall situation of the group members in the vehicle by providing appropriate reference information corresponding to the situation in a timely manner through a conversational agent system. Manufacturing a prototype of the system, we verified both its operations involved in the test case and its capability to infer the intentions and desires of each group member and intervene to their conversation in an appropriate timing. In this research, we have demonstrated that our method used for the prototype was appropriate to a practical use through. In future, we will focus on optimizing the logic and system functions of group situation estimation and the subsequent steps of providing the reference information.
CITATION STYLE
Kono, S., Wakisaka, Y., & Ikeno, A. (2016). Prototype of conversation support system for activating group conversation in the vehicle. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9733, pp. 119–127). Springer Verlag. https://doi.org/10.1007/978-3-319-39513-5_11
Mendeley helps you to discover research relevant for your work.