| 摘要 |
PROBLEM TO BE SOLVED: To provide a configuration which sets a projection destination microphone which differs from a sound source separation microphone and performs projection processings. SOLUTION: Mixed signals are subjected to separation processing, by applying independent component analysis (ICA) for observation signals, based on mixed signals of two or more sound sources obtained by a sound source separation microphone and separation signals corresponding to each sound source are generated. Next, generated separation signals and observation signals of a projection destination microphone which differ from the sound source separation microphone are input, and projection signals are generated which are separation signals corresponding to each sound source assumed to be obtained by the projection destination microphone by applying these input signals. Furthermore, it is made possible for projection signals to output voice data for an output device and to infer the direction or the location of the sound source. COPYRIGHT: (C)2011,JPO&INPIT
|
