Unifying Local and Global Methods for Harmonic-Percussive Source Separation

• Published in: 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) pp. 176 - 180
• Main Authors: Dittmar, Christian, Lopez-Serrano, Patricio, Muller, Meinard
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.04.2018
• Subjects: Harmonic analysis; harmonic percussive source separation; Instruments; Kernel; music decomposition; Signal processing algorithms; signal reconstruction; Source separation; Spectrogram; Task analysis
• ISSN: 2379-190X
• DOI: 10.1109/ICASSP.2018.8462119

This paper addresses the separation of drums from music recordings, a task closely related to harmonic-percussive source separation (HPSS). In previous works, two families of algorithms have been prominently applied to this problem. They are based either on local filtering and diffusion schemes, or on global low-rank models. In this paper, we propose to combine the advantages of both paradigms. To this end, we use a local approach based on Kernel Additive Modeling (KAM) to extract an initial guess for the percussive and harmonic parts. Subsequently, we use Non-Negative Matrix Factorization (NMF) with soft activation constraints as a global approach to jointly enhance both estimates. As an additional contribution, we introduce a novel constraint for enhancing percussive activations and a scheme for estimating the percussive weight of NMF components. Throughout the paper, we use a real-world music example to illustrate the ideas behind our proposed method. Finally, we report promising BSS Eval results achieved with the publicly available test corpora ENST-Drums and QUASI, which contain isolated drum and accompaniment tracks.