Sound Source Separation using Shifted Non-negative Tensor Factorisation

Authors

Eugene Coyle, Technological University DublinFollow
Derry Fitzgerald, Technological University DublinFollow
Matt Cranitch, Technological University Dublin

Document Type

Conference Paper

Rights

This item is available under a Creative Commons License for non-commercial use only

Publication Details

In IEEE International Conference on Acoustics, Speech and Signal Processing, Toulouse, France, 2006 http://www.audioresearchgroup.com/

Abstract

Recently, shifted non-negative Matrix Factorisation was developed as a means of separating harmonic instruments from single channel mixtures. However, in many cases two or more channels are available, in which case it would be advantageous to have a multichannel version of the algorithm. To this end, a shifted Non-negative Tensor Factorisation algorithm is derived, which extends shifted Non-negative Matrix Factoristiaon to the multi channel case. The use of this algorithm for multi-channel sound source separation of harmonic instruments is demonstrated. Further, it is shown that the algorithm can be used to perform Non-negative Tensor Deconvolution, to separate sound sources which have time evolving spectra from multi-channel signals.

Recommended Citation

Coyle, E., Fitzgerald, D. & Cranitch, M.: Sound source separation using shifted non-negative tensor factorisation. IEEE International Conference on Acoustics, Speech and Signal Processing, Toulouse, France, 2006.