Opens in a new tab

AES E-Library

← Back to search

Conference Paper

Toward Live Drum Separation Using Probabilistic Spectral Clustering Based on the Itakura-Saito Divergence

Authors: Battenberg, Eric; Huang, Victor; Wessel, David

AES Conference: 45th International Conference: Applications of Time-Frequency Processing in Audio · Paper 1-2 · March 2012

Abstract

We present a live drum separation system for a specific target drumset to be used as a front end in a complete live drum understanding system. Our system decomposes drum note onsets onto spectral drum templates by adapting techniques from non-negative matrix factorization. Multiple templates per drum are computed using a new gamma mixture model clustering procedure to account for the variety of sounds that can be produced by a single drum. This clustering procedure imposes an Itakura-Saito distance metric on the cluster space. In addition, we utilize “tail” templates for each drum which greatly improve the separation accuracy when cymbals with long decay times are present.

Details

Published in
AES Conference: 45th International Conference: Applications of Time-Frequency Processing in Audio
Paper number
1-2
Publication date
March 6, 2012
Session subject
Processing of Audio
Affiliation
University of California, Berkeley, CA, USA; Center for New Music and Audio Technologies, Berkeley, CA, USA (See document for exact affiliation information.)
Type
Conference Paper