摘要 |
<P>PROBLEM TO BE SOLVED: To separate each signal component and assign each signal component to a proper channel, even when each signal component included in an input audio signal is not orthogonal. <P>SOLUTION: The input audio signals of each one of right and left channels each of which expresses mixed sound of sound emitted from each of at least two sound sources are acquired, and frequency domain independent component analysis is performed to extract signal components corresponding to the sound of each sound source. Then, one of the signal components is removed from each input audio signal to generate a first output audio signal of each one of the right and left channels, and the other signal component is removed from each input audio signal to generate a second output audio signal of each one of the right and left channels. <P>COPYRIGHT: (C)2011,JPO&INPIT |