[Japanese | Thesis | Researches in Minoh Lab | Minoh Lab]
This paper proposes an audio telecommunication method that allows users to attend multiple meetings.
It is getting more and more important to support informal communication. It is generally pointed out that many conservative telecommunication tools do not support users to find out a meeting space where they can get interesting information. In conservative telecommunication tools, users have to visit meeting spaces one by one in order to find interesting information which exists in many meeting spaces.
We propose a teleconference system which support users to attend multiple meetings at the same time. In this system, users can get information from multiple meetings at the same time, and can more likely to get interesting information in comparison with the system that support users to attend only one meeting.
We construct the system using audio media because it is easy to use and includes non-verbal information, and these two points are essential for informal communication.
To construct the system, we must achieve two goals. One is ``to allow users to get information from multiple meetings at the same time,'' and the other is ``to avoid multiple audio muddling.'' These two goals have a trade off relationship. In the system which provides users audio from multiple meetings at the same time (achieving the former goal), it is difficult to avoid multiple audio muddling. On the other hand, in the system which provides users audio from only one meeting, users can avoid audio muddling (achieving the latter goal), but cannot listen to the audio from other meetings, and as a result they lose chances to get interesting information.
To solve this trade off problem, we propose an audio media controlling method based on the users' intention to get information in each meeting. In this paper, users' intention is defined as the degree of intensity users have on each meeting.
We define the concept of intention by supposing the situation we join multiple conversations in real world. When we listen to multiple conversations in real world, we do not listen to all conversations completely. But we can understand multiple conversations by trying to get an outline of a topic we are not interested in, and get detailed information of a topic we are interested in. So, when a user needs an outline of a topic in a meeting, we define his or her intention to the meeting as low degree, and when a user needs detailed information of a topic in a meeting, we define his or her intention to the meeting as high degree.
Based on the consideration, we propose a method to attend multiple meetings by audio control according to users' intention, and implement the system. In this paper, we conducted an experiment of attending multiple meetings by the system. The result shows the proposed method enable users to attend multiple meetings at the same time without feeling audio muddling.