BLIND SOURCE SEPARATION USING REPETITIVE STRUCTURE

Blind source separation algorithms typically involve decorrelat- ing time-aligned mixture signals. The usual assumption is that all sources are active at all times. However, if this is not the case, we show that the unique pattern of source activity/inactivity helps sep- aration. Music is the most obvious example of sources exhibiting repetitive structure because it is carefully constructed. We present a novel source separation algorithm based on spatial time-time dis- tributions that capture the repetitive structure in audio. Our method outperforms time-frequency source separation when source spec- tra are highly overlapping.