A Multiplicative Algorithm for Convolutive Non-Negative Matrix Factorization Based on Squared Euclidean Distance

Using the convolutive nonnegative matrix factorization (NMF) model due to Smaragdis, we develop a novel algorithm for matrix decomposition based on the squared Euclidean distance criterion. The algorithm features new formally derived learning rules and an efficient update for the reconstructed nonnegative matrix. Performance comparisons in terms of computational load and audio onset detection accuracy indicate the advantage of the Euclidean distance criterion over the Kullback-Leibler divergence criterion.

[1]  Irfan A. Essa,et al.  Incorporating Phase Information for Source Separation via Spectrogram Factorization , 2007, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07.

[2]  Christiane Quesne,et al.  Linear Canonical Transformations and Their Unitary Representations , 1971 .

[3]  P. Smaragdis,et al.  Non-negative matrix factorization for polyphonic music transcription , 2003, 2003 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (IEEE Cat. No.03TH8684).

[4]  Kristoffer Hougaard Madsen,et al.  Shifted Non-Negative Matrix Factorization , 2007, 2007 IEEE Workshop on Machine Learning for Signal Processing.

[5]  Mark B. Sandler,et al.  A tutorial on onset detection in music signals , 2005, IEEE Transactions on Speech and Audio Processing.

[6]  Haldun M. Özaktas,et al.  Fractional Fourier domains , 1995, Signal Process..

[7]  Billur Barshan,et al.  Optimal filtering with linear canonical transformations , 1997 .

[8]  Soo-Chang Pei,et al.  Eigenfunctions of linear canonical transform , 2002, IEEE Trans. Signal Process..

[9]  Luís B. Almeida,et al.  The fractional Fourier transform and time-frequency representations , 1994, IEEE Trans. Signal Process..

[10]  Tuomas Virtanen,et al.  Monaural Sound Source Separation by Nonnegative Matrix Factorization With Temporal Continuity and Sparseness Criteria , 2007, IEEE Transactions on Audio, Speech, and Language Processing.

[11]  Patrik O. Hoyer,et al.  Non-negative Matrix Factorization with Sparseness Constraints , 2004, J. Mach. Learn. Res..

[12]  Martin J. Bastiaans,et al.  On fractional Fourier transform moments , 2000, IEEE Signal Processing Letters.

[13]  L. Bernardo.  ABCD MATRIX FORMALISM OF FRACTIONAL FOURIER OPTICS , 1996 .

[14]  Adrian Stern,et al.  Uncertainty principles in linear canonical transform domains and some of their implications in optics. , 2008, Journal of the Optical Society of America. A, Optics, image science, and vision.

[15]  Vikram M. Gadre,et al.  An uncertainty principle for real signals in the fractional Fourier transform domain , 2001, IEEE Trans. Signal Process..

[16]  Paris Smaragdis,et al.  Non-negative Matrix Factor Deconvolution; Extraction of Multiple Sound Sources from Monophonic Inputs , 2004, ICA.

[17]  Queen Mary.  MUSICAL AUDIO STREAM SEPARATION BY NON-NEGATIVE MATRIX FACTORIZATION , 2005 .

[18]  Derry Fitzgerald,et al.  Sound Source Separation Using Shifted Non-Negative Tensor Factorisation , 2006, 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.

[19]  Kamalesh Kumar Sharma,et al.  Signal separation using linear canonical and fractional Fourier transforms , 2006 .

[20]  Andrzej Cichocki,et al.  Csiszár's Divergences for Non-negative Matrix Factorization: Family of New Algorithms , 2006, ICA.

[21]  C. Capus,et al.  Short-time fractional Fourier methods for the time-frequency representation of chirp signals. , 2003, The Journal of the Acoustical Society of America.

[22]  Wenwu Wang,et al.  Non-Negative Matrix Factorization for Note Onset Detection of Audio Signals , 2006, 2006 16th IEEE Signal Processing Society Workshop on Machine Learning for Signal Processing.

[23]  Wenwu Wang,et al.  Squared Euclidean Distance Based Convolutive Non-Negative Matrix Factorization with Multiplicative Learning Rules For Audio Pattern Separation , 2007, 2007 IEEE International Symposium on Signal Processing and Information Technology.

[24]  Paris Smaragdis,et al.  Convolutive Speech Bases and Their Application to Supervised Speech Separation , 2007, IEEE Transactions on Audio, Speech, and Language Processing.

[25]  Kamalesh Kumar Sharma,et al.  Uncertainty Principle for Real Signals in the Linear Canonical Transform Domains , 2008, IEEE Transactions on Signal Processing.

[26]  H. Sebastian Seung,et al.  Algorithms for Non-negative Matrix Factorization , 2000, NIPS.

[27]  Michael A. Casey,et al.  Separation of Mixed Audio Sources By Independent Subspace Analysis , 2000, ICMC.

[28]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[29]  Barak A. Pearlmutter,et al.  Convolutive Non-Negative Matrix Factorisation with a Sparseness Constraint , 2006, 2006 16th IEEE Signal Processing Society Workshop on Machine Learning for Signal Processing.

[30]  Morten Mørup,et al.  Nonnegative Matrix Factor 2-D Deconvolution for Blind Single Channel Source Separation , 2006, ICA.

[31]  Mikkel N. Schmidt,et al.  Sparse Non-negative Matrix Factor 2-D Deconvolution , 2006 .