摘要 |
PROBLEM TO BE SOLVED: To facilitate discrimination of a conversation state by making the discrimination, based on only a speech of a leader in the conversation. SOLUTION: A conversation state discriminating device 10 includes a computer 12, which receives and inputs the output of a microphone 14 to a database 16. Namely, a speech signal, corresponding to the utterance of the leader A of the conversation, is registered in the database 16. The computer 12 detects an utterance section and a voiceless section from the speech signal and then discriminates the conversation state, based on the length of the voiceless section. Namely, the conversation state at a place where the leader A of the conversation is present is discriminated. COPYRIGHT: (C)2005,JPO&NCIPI
|