E. Battenberg, V. Huang, and D. Wessel, “Toward Live Drum Separation Using Probabilistic Spectral Clustering Based on the Itakura-Saito Divergence,” in Proc. AES Conference: 45th International Conference: Applications of Time-Frequency Processing in Audio, Mar. 2012, Paper 1-2. [Online]. Available: https://aes.org/publications/elibrary-page/?id=16176
Battenberg E, Huang V, Wessel D. Toward Live Drum Separation Using Probabilistic Spectral Clustering Based on the Itakura-Saito Divergence. In: AES Conference: 45th International Conference: Applications of Time-Frequency Processing in Audio. Audio Engineering Society; 2012. Paper 1-2. Available from: https://aes.org/publications/elibrary-page/?id=16176
@inproceedings{Battenberg2012_16176,
author = {Battenberg, Eric and Huang, Victor and Wessel, David},
title = {{Toward Live Drum Separation Using Probabilistic Spectral Clustering Based on the Itakura-Saito Divergence}},
booktitle = {AES Conference: 45th International Conference: Applications of Time-Frequency Processing in Audio},
note = {Paper 1-2},
year = {2012},
month = mar,
publisher = {Audio Engineering Society},
url = {https://aes.org/publications/elibrary-page/?id=16176}
}
TY - CPAPER
TI - Toward Live Drum Separation Using Probabilistic Spectral Clustering Based on the Itakura-Saito Divergence
AU - Battenberg, Eric
AU - Huang, Victor
AU - Wessel, David
T2 - AES Conference: 45th International Conference: Applications of Time-Frequency Processing in Audio
M1 - Paper 1-2
PY - 2012
DA - 2012/03/06
UR - https://aes.org/publications/elibrary-page/?id=16176
PB - Audio Engineering Society
LA - en
AB - 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.
ER -