Emotion recognition and its application to computer agents with spontaneous interactive capabilities

In this paper, we first study the recognition of emotions involved in human speech. We propose an emotion recognition algorithm based on a neural netw ork and also propose a method to collect a large speech database that conta ins emotions. We carried out emotion recognition experiments based on the n eural network trained using this database. An emotion recognition rate of a pproximately 50% was obtained in a speaker-independent mode for eight emoti on states. We then tried to apply this emotion recognition algorithm to a computer age nt that plays a character role in the interactive movie system we are devel oping. We propose to use emotion recognition as key technology for an archi tecture of the computer characters with both narrative-based and spontaneou s interaction capabilities. (C) 2000 Elsevier Science B.V. All rights reser ved.