SOUND SOURCE SEPARATION SYSTEM, SOUND SOURCE POSITION ESTIMATION SYSTEM, SOUND SOURCE SEPARATION METHOD, AND SOUND SOURCE SEPARATION PROGRAM

摘要

[Problem] To provide a technology that enables efficient use of a characteristic sound collection result in a case where a sound source exits at a position apart from a plurality of microphones at equal distances. [Solution] To constitute a sound source separation system comprising: a sound collection data acquisition unit that acquires sound collection data obtained by collecting sounds output from a first sound source exiting at a first position apart from the plurality of microphones at an equal effective distance and a second sound source existing at a position other than the first position by the plurality of microphones; and a frequency spectrum acquisition unit that acquires, on the basis of the sound collection data, a frequency spectrum as to two dimensions of a circumferential direction of a circle, the center of which is the first position and the radius of which is the effective distance and a time direction; and a sound source separation unit that separates a first sound source spectrum as a spectrum of a sound output from the first sound source and a second sound source spectrum as a spectrum of a sound output from the second sound source on the basis of the frequency spectrum.