Sparse Non-negative Tensor 2D Deconvolution (SNTF2D) for multi channel time-frequency analysis

Morten Mørup; Mikkel N. Schmidt

Sparse Non-negative Tensor 2D Deconvolution (SNTF2D) for multi channel time-frequency analysis

Research output: Book/Report › Report › Research › peer-review

118 Downloads (Pure)

Abstract

We recently introduced two algorithms for sparse non-negative matrix factor 2-D deconvolution (SNMF2D) that are useful for single channel source separation and music transcription. We here extend this approach to the analysis of the log-frequency spectrograms of a multichannel recording. The model proposed forms a non-negative tensor factor 2-D deconvolution (NTF2D) based on the parallel factor (PARAFAC) model. Two algorithms are given for NTF2D; one based on least squares the other on Kullback-Leibler divergence minimization. Both algorithms are extended to give sparse decompositions. The algorithms are demonstrated to successfully identify the components of both artificially generated as well as real stereo music.

Original language	English

Publication status	Published - 2006

Access to Document

Imm4659Final published version, 497 KB

Fingerprint Dive into the research topics of 'Sparse Non-negative Tensor 2D Deconvolution (SNTF2D) for multi channel time-frequency analysis'. Together they form a unique fingerprint.

Cite this

@book{095bba863e464559aec57dac8e6c4db9,

title = "Sparse Non-negative Tensor 2D Deconvolution (SNTF2D) for multi channel time-frequency analysis",

abstract = "We recently introduced two algorithms for sparse non-negative matrix factor 2-D deconvolution (SNMF2D) that are useful for single channel source separation and music transcription. We here extend this approach to the analysis of the log-frequency spectrograms of a multichannel recording. The model proposed forms a non-negative tensor factor 2-D deconvolution (NTF2D) based on the parallel factor (PARAFAC) model. Two algorithms are given for NTF2D; one based on least squares the other on Kullback-Leibler divergence minimization. Both algorithms are extended to give sparse decompositions. The algorithms are demonstrated to successfully identify the components of both artificially generated as well as real stereo music.",

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

year = "2006",

language = "English",

}

TY - RPRT

T1 - Sparse Non-negative Tensor 2D Deconvolution (SNTF2D) for multi channel time-frequency analysis

AU - Mørup, Morten

AU - Schmidt, Mikkel N.

PY - 2006

Y1 - 2006

N2 - We recently introduced two algorithms for sparse non-negative matrix factor 2-D deconvolution (SNMF2D) that are useful for single channel source separation and music transcription. We here extend this approach to the analysis of the log-frequency spectrograms of a multichannel recording. The model proposed forms a non-negative tensor factor 2-D deconvolution (NTF2D) based on the parallel factor (PARAFAC) model. Two algorithms are given for NTF2D; one based on least squares the other on Kullback-Leibler divergence minimization. Both algorithms are extended to give sparse decompositions. The algorithms are demonstrated to successfully identify the components of both artificially generated as well as real stereo music.

AB - We recently introduced two algorithms for sparse non-negative matrix factor 2-D deconvolution (SNMF2D) that are useful for single channel source separation and music transcription. We here extend this approach to the analysis of the log-frequency spectrograms of a multichannel recording. The model proposed forms a non-negative tensor factor 2-D deconvolution (NTF2D) based on the parallel factor (PARAFAC) model. Two algorithms are given for NTF2D; one based on least squares the other on Kullback-Leibler divergence minimization. Both algorithms are extended to give sparse decompositions. The algorithms are demonstrated to successfully identify the components of both artificially generated as well as real stereo music.

M3 - Report

BT - Sparse Non-negative Tensor 2D Deconvolution (SNTF2D) for multi channel time-frequency analysis

ER -