Maynooth University

Maynooth University ePrints and eTheses Archive

Maynooth University Library

Drum Source Separation using Percussive Feature Detection and Spectral Modulation

UNSPECIFIED (2005) Drum Source Separation using Percussive Feature Detection and Spectral Modulation. In: UNSPECIFIED.

[img] Download (127kB)

Abstract

We present a method for the separation and resynthesis of drum sources from single channel polyphonic mixtures. The frequency domain technique involves identifying the presence of a drum using a novel percussive feature detection function, after which the short-time magnitude spectrum is estimated and scaled according to an estimated time-amplitude function derived from the percussive measure. In addition to producing high quality separation results, the method we describe is also a useful pre-process for drum transcription techniques such as Prior Subspace Analysis in the presence of pitched instruments.

Item Type: Conference or Workshop Item (Other)
Additional Information: This paper is a postprint of a paper submitted to and accepted for publicatin in (journal/conference) and is subject to Institution of Engineering and Technology Copyright. The copy of record is available at IET Digital Library
Keywords: Audio, Signal Processing, Source Separation, Drum.
Subjects: Science & Engineering > Electronic Engineering
Item ID: 699
Depositing User: Robert Lawlor
Date Deposited: 06 Sep 2007
URI:

Repository Staff Only(login required)

View Item Item control page

Document Downloads

More statistics for this item...