摘要 |
<P>PROBLEM TO BE SOLVED: To acquire a voice by voice acquisition means to estimate a psychological state of a speaker on the basis of nonverbal information of the acquired voice. <P>SOLUTION: A voice analysis system includes terminal devices each of which is worn on a user, and a host device which acquires information from the terminal devices. Each terminal device analyzes a voice acquired by a microphone to identify a speaker and estimates a level of mental activity of the speaker on the basis of a voice feature quantity being nonverbal information and transmits utterance information including the estimation result to the host device. The host device extracts utterance information relating to a specific conversation, from utterance information acquired from a plurality of terminal devices and analyzes the utterance information and estimates emotion of each of speakers participating in the conversation on the basis of combination of levels of mental activity of respective speakers. <P>COPYRIGHT: (C)2013,JPO&INPIT |