ePrints

Underdetermined Convolutive Source Separation using GEM-MU with Variational Approximated Optimum Model Order NMF2D

Lookup NU author(s): Dr Wai Lok Woo, Emeritus Professor Satnam Dlay, Dr Bin Gao

Downloads

• Accepted version [.pdf]

Licence

This is the authors' accepted manuscript of an article that has been published in its final definitive form by IEEE, 2017.

For re-use rights please refer to the publisher's terms and conditions.

Abstract

An unsupervised machine learning algorithm based on nonnegative matrix factor 2D deconvolution (NMF2D) with approximated optimum model order is proposed. The proposed algorithm adapted under the hybrid framework that combines the generalized EM algorithm with multiplicative update (GEM-MU). As the number of parameters in the NMF2D grows exponentially as the number of frequency basis increases linearly, the issues of model order fitness, initialization and parameters estimation become ever more critical. This paper proposes a variational Bayesian method to optimize the number of components in the NMF2D by using the Gamma-Exponential process as the observation-latent model. In addition, it is shown that the proposed Gamma-Exponential process can be used to initialize the NMF2D parameters. Finally, the paper investigates the issue and advantages of using different window length. Experimental results for the synthetic convolutive mixtures and live recordings verify the competence of the proposed algorithm.

Publication metadata

Author(s): Al-Tmeme A, Woo WL, Dlay SS, Gao B

Publication type: Article

Publication status: Published

Journal: IEEE/ACM Transactions on Audio, Speech and Language Processing

Year: 2017

Volume: 25

Issue: 1

Pages: 35-49

Print publication date: 01/01/2017

Online publication date: 24/10/2016

Acceptance date: 14/10/2016

Date deposited: 17/10/2016

ISSN (print): 2329-9290

ISSN (electronic): 2329-9304

Publisher: IEEE

URL: http://dx.doi.org/10.1109/TASLP.2016.2620600

DOI: 10.1109/TASLP.2016.2620600

Altmetrics

Share