摘要 |
PROBLEM TO BE SOLVED: To provide a sound source separation device capable of efficiently performing a sound source separation even in a dispersal microphone array environment.SOLUTION: A microphone separate sound source existence posterior probability estimation part estimates a sound source existence posterior probability about each sound source for each microphone using observation signals of plural channels in which sound source signals emitted from a plurality of sound sources are collected by a plurality of microphones and a model of the observation signal which assumes that a sound pressure of a signal from each of the plurality of sound sources observed by each of the plurality of microphones is different. A model parameter estimation part estimates a model parameter of the observation signal with the observation signals of the plural channels and the sound source existence posterior probability as an input. An output sound estimation part estimates and outputs an arrival signal about each sound source for each microphone with the observation signals of the plural channels, the sound source existence posterior probability, and the model parameter as an input. |