Nonnegative Matrix Factor 2-D Deconvolution for Blind Single Channel Source Separation

Mikkel N. Schmidt; Morten Mørup

Nonnegative Matrix Factor 2-D Deconvolution for Blind Single Channel Source Separation

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

587 Downloads (Pure)

Abstract

We present a novel method for blind separation of instruments in polyphonic music based on a non-negative matrix factor 2-D deconvolution algorithm. Using a model which is convolutive in both time and frequency we factorize a spectrogram representation of music into components corresponding to individual instruments. Based on this factorization we separate the instruments using spectrogram masking. The proposed algorithm has applications in computational auditory scene analysis, music information retrieval, and automatic music transcription.

Original language	English
Title of host publication	ICA2006 : Source Separation and Independent Component Analysis, International Conference on (ICA)
Publication date	2006
Publication status	Published - 2006
Event	Source Separation and Independent Component Analysis, International Conference on (ICA) - Duration: 1 Jan 2006 → …

Conference

Conference	Source Separation and Independent Component Analysis, International Conference on (ICA)
Period	01/01/2006 → …

Access to Document

Imm4061Final published version, 171 KB

http://www2.imm.dtu.dk/pubdb/p.php?4061

OpenUrl availability

Full text

Cite this

@inproceedings{dcb70074483c44e0b8575973bbd0c446,

title = "Nonnegative Matrix Factor 2-D Deconvolution for Blind Single Channel Source Separation",

abstract = "We present a novel method for blind separation of instruments in polyphonic music based on a non-negative matrix factor 2-D deconvolution algorithm. Using a model which is convolutive in both time and frequency we factorize a spectrogram representation of music into components corresponding to individual instruments. Based on this factorization we separate the instruments using spectrogram masking. The proposed algorithm has applications in computational auditory scene analysis, music information retrieval, and automatic music transcription.",

keywords = "non-negative matrix factorization, blind source separation",

author = "Schmidt, {Mikkel N.} and Morten M{\o}rup",

year = "2006",

language = "English",

booktitle = "ICA2006",

note = "Source Separation and Independent Component Analysis, International Conference on (ICA) ; Conference date: 01-01-2006",

}

TY - GEN

T1 - Nonnegative Matrix Factor 2-D Deconvolution for Blind Single Channel Source Separation

AU - Schmidt, Mikkel N.

AU - Mørup, Morten

PY - 2006

Y1 - 2006

N2 - We present a novel method for blind separation of instruments in polyphonic music based on a non-negative matrix factor 2-D deconvolution algorithm. Using a model which is convolutive in both time and frequency we factorize a spectrogram representation of music into components corresponding to individual instruments. Based on this factorization we separate the instruments using spectrogram masking. The proposed algorithm has applications in computational auditory scene analysis, music information retrieval, and automatic music transcription.

AB - We present a novel method for blind separation of instruments in polyphonic music based on a non-negative matrix factor 2-D deconvolution algorithm. Using a model which is convolutive in both time and frequency we factorize a spectrogram representation of music into components corresponding to individual instruments. Based on this factorization we separate the instruments using spectrogram masking. The proposed algorithm has applications in computational auditory scene analysis, music information retrieval, and automatic music transcription.

KW - non-negative matrix factorization

KW - blind source separation

M3 - Article in proceedings

BT - ICA2006

T2 - Source Separation and Independent Component Analysis, International Conference on (ICA)

Y2 - 1 January 2006

ER -