A method for determining a talk/listen state using voice detection includes receiving an audio sample and detecting whether the audio sample includes voiced sound. The audio sample represents sound measured during a sample time interval. The method further includes deriving an audio level from the audio...http://www.google.fr/patents/US5991277?utm_source=gb-gplus-shareBrevet US5991277 - Primary transmission site switching in a multipoint videoconference environment based on human voice
Primary transmission site switching in a multipoint videoconference ...
