OKI and a group of researchers of Waseda University in Tokyo have developed a device that is able to extract a person’s voice from a mix of background noise and other persons’ voices. The new voice-extraction technology is combined with a voice-recognition system that makes it possible to control appliances via spoken commands.
The prototype (which there are no photos of, I looked everywhere on the Japanese web) supposedly has the size and look of a typical remote control with four small microphones attached to its front. Users can dim the lights, turn on electric appliances or change TV programs by speaking into the device.
The technology is also suitable for recording lectures or speeches, cutting the voices of other people present and general background noise.
Via Nikkei [registration required / paid subscription]