Document Type



This item is available under a Creative Commons License for non-commercial use only


2.2 ELECTRICAL, ELECTRONIC, INFORMATION ENGINEERING, Electrical and electronic engineering

Publication Details

Electronics 2020,xx, 5; doi:10.3390/electronicsxx010005


Binary masking forms the basis for a number of source separation approaches that havebeen successfully applied to the problem of de-mixing music sources from a stereo recording.A well-known problem with binary masking is that, when music sources overlap in the time-frequencydomain, only one of the overlapping sources can be assigned the energy in a particulartime-frequency bin. To overcome this problem, we reformulate the classical pan-pot source separationproblem for music sources as a non-negative quadratic program. This reformulation gives rise toan algorithm, called Redress, which extends the popular Adress algorithm. It works by definingan azimuth trajectory for each source based on its spatial position within the stereo field. Redressallows for the allocation of energy in one time-frequency bin to multiple sources. We present resultsthat show that for music recordings Redress improves the SNR, SAR, and SDR in comparison to theAdress algorithm.