Toward the Human Communication Efficiency Monitoring from Captured Audio and Video Media in Real Environments
The paper presents a method designed to monitor the human–to–human communication and to estimate its efficiency based on recorded audio and video. The efficiency is estimated during the communication process, based on face and body movements tracking and nonverbal audio information classification. The members of the communication process are temporally classified as senders and receivers for short time intervals, since the role in the process usually changes in time. For our experiments we consider the situations of communication that is intended by their members (intentional) and that can be observed as main activity (primary communication) like project meetings, lectures or academic debates, where people join the discussion and the direction of information (knowledge) flow can be evaluated.
KeywordsHuman Communication Nonnegative Matrix Factorization Speech Recognizer Primary Communication Audio Information
Unable to display preview. Download preview PDF.
- 1.Itou, J., Kakusho, K., Minoh, M.: Observing Human Faces in Communication to Understand the Atmosphere. In: Proceedings of 7th Pacific Rim International Conference on Artificial Intelligence, Tokyo, pp. 30–33 (2002)Google Scholar
- 2.King, R.G.: Fundamentals of Human Communication. Macmillan Publishing Co., Inc., New York (1979)Google Scholar
- 5.Hershey, J., Movellan, J.: Audio-Vision: Using Audio-Visual Synchrony to Locate Sounds. In: Solla, S.A., Leen, T.K., Mller, K.-R. (eds.) Neural Information Processing Systems 12, MIT Press, Cambridge (2000)Google Scholar