Drum stroke detection and recognition

Jukka Sillanpää,   s157529@cs.tut.fi,  +358-3-3652124
Tampere University of Technology
Signal Processing Laboratory
P.O.Box 553, FIN-33101 Tampere, Finland

Introduction

The goal of drum stroke detection and recognition is to extract all the drum information from musical signal. In other words task of the system is to find out the onset moments of drum sounds and further, recognize the drum or drum combination that caused each of these drum sounds. To do that the system observes the frequency contents altering of musical signal. How the observation is done and what is the recognition algorithm like is explained in the following document.

Documents

Sillanpää, Klapuri, Seppänen, Virtanen. (2000). " Recognition of acoustic noise mixtures by combined bottom-up and top-down processing ". In proc. European Signal Processing Conference, EUSIPCO 2000.

Sillanpää. (2000). " Drum stroke detection and recognition ".

Example

As an example of frequency altering observation we have a music sample sample.wav and a freguency shape matrix "sample" formed from the musical signal.


Last modified: Fri Sep 29 12:09:05 EET DST 2000 - Jukka Sillanpää s157529@cs.tut.fi