Non-negative tensor factorisation of modulation spectrograms for monaural sound source separation

This paper proposes an algorithm for separating monaural audio signals by non-negative tensor factorisation of modulation spectrograms. The modulation spectrogram is able to represent redundant patterns across frequency with similar features, and the tensor factorisation is able to isolate these patterns in an unsupervised way. The method overcomes the limitation of conventional non-negative matrix factorisation algorithms to utilise the redundancy of sounds in frequency. In the proposed method, separated sounds are synthesised by filtering the mixture signal with a Wiener-like filter generated from the estimated tensor factors. The proposed method was compared to conventional algorithms in unsupervised separation of mixtures of speech and music. Improved signal to distortion ratios were obtained compared to standard non-negative matrix factorisation and non-negative matrix deconvolution.

[1]  Steven Greenberg,et al.  The modulation spectrogram: in pursuit of an invariant representation of speech , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[2]  D. Fitzgerald,et al.  Using Tensor Factorisation Models to Separate Drums from Polyphonic Music , 2009 .

[3]  D. Fitzgerald,et al.  Non-negative Tensor Factorisation for Sound Source Separation , 2005 .

[4]  R. Patterson,et al.  Complex Sounds and Auditory Images , 1992 .

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

[6]  Christopher J. Plack,et al.  The Sense of Hearing , 2005 .

[7]  Björn W. Schuller,et al.  Optimization and Parallelization of Monaural Source Separation Algorithms in the openBliSSART Toolkit , 2012, J. Signal Process. Syst..

[8]  Emmanuel Vincent,et al.  Enforcing Harmonicity and Smoothness in Bayesian Non-Negative Matrix Factorization Applied to Polyphonic Music Transcription , 2010, IEEE Transactions on Audio, Speech, and Language Processing.

[9]  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.

[10]  Rémi Gribonval,et al.  Performance measurement in blind audio source separation , 2006, IEEE Transactions on Audio, Speech, and Language Processing.

[11]  Derry Fitzgerald,et al.  Extended Nonnegative Tensor Factorisation Models for Musical Sound Source Separation , 2008, Comput. Intell. Neurosci..

[12]  Peter Ladefoged,et al.  On the Fusion of Sounds Reaching Different Sense Organs , 1957 .

[13]  A. Klapuri,et al.  Analysis of polyphonic audio using source-filter model and non-negative matrix factorization , 2006 .

[14]  Anssi Klapuri,et al.  Multipitch Analysis of Polyphonic Music and Speech Signals Using an Auditory Model , 2008, IEEE Transactions on Audio, Speech, and Language Processing.

[15]  Masataka Goto,et al.  RWC Music Database: Popular, Classical and Jazz Music Databases , 2002, ISMIR.

[16]  Bhiksha Raj,et al.  Non-negative matrix factorization based compensation of music for automatic speech recognition , 2010, INTERSPEECH.