Why does music source separation benefit from cacophony?
Author:
Chang-Bin Jeon, Gordon Wichern, François G. Germain, Jonathan Le Roux
Keyword:
Electrical Engineering and Systems Science, Audio and Speech Processing, Audio and Speech Processing (eess.AS)
journal:
--
date:
2024-02-28 00:00:00
Abstract
In music source separation, a standard training data augmentation procedure is to create new training samples by randomly combining instrument stems from different songs. These random mixes have mismatched characteristics compared to real music, e.g., the different stems do not have consistent beat or tonality, resulting in a cacophony. In this work, we investigate why random mixing is effective when training a state-of-the-art music source separation model in spite of the apparent distribution shift it creates. Additionally, we examine why performance levels off despite potentially limitless combinations, and examine the sensitivity of music source separation performance to differences in beat and tonality of the instrumental sources in a mixture.