BLIND RHYTHMIC SOURCE SEPARATION: NONNEGATIVITY AND REPEATABILITY

An unsupervised method is proposed aiming at extracting rhythmic sources from commercial polyphonic music whose number of chan-nels is limited to one. Commercial music signals are not usually provided with more than two channels while they often contain mul-tiple instruments including singing voice. Therefore, instead of us-ing conventional ways, such as modeling mixing environments or statistical characteristics, we should introduce other source-specific characteristics for separating or extracting the sources. In this pa-per, we concentrate on extracting rhythmic sources from the mixture with the other harmonic sources. An extension of nonnegative ma-trix factorization (NMF) is used to analyze multiple relationships between spectral and temporal properties in the given input matri-ces. Moreover, temporal repeatability of the rhythmic sound sources is implicated as common rhythmic property among segments of an input mixture signal. The proposed method shows acceptable, but not superior separation quality to the referred drum source separa-tion systems. However, it has better applicability due to its blind manner in separation. Index Terms — Nonnegative matrix factorization, rhythmic source separation, musical information researc