| 摘要 |
<P>PROBLEM TO BE SOLVED: To allow participants in a teleconference to easily know who is speaking, who is listening, and who starts speaking when. <P>SOLUTION: Background noise data unique to each conference participant is generated, and background noise data of a speaker among the conference participants is combined with data of speech sound from the speaker for reproduction. Nodding movement, hand movement and head movement of listeners are detected. <P>COPYRIGHT: (C)2012,JPO&INPIT |
