Title of Paper:Single-Channel Blind Separation using L1-Sparse Complex Nonnegative Matrix Factorization for Acoustic Signals

Hits:

Journal:Journal of the Acoustical Society of America

Abstract:A novel single-channel audio source separation (SCASS) has been developed to extract better quality of audio separated signals. This approach will exploit the variational -sparse complex matrix factorization (v-SCMF) which offers the advantages of the CMF and a variational -sparse approach simultaneously. CMF is based on a mixing model defined in the complex-spectrum domain and estimates recurring patterns in the observed magnitude spectra, their activations and their phases. The proposed factorization decomposes an information-bearing matrix into complex of factor matrices.

All the Authors: Gao B, Dlay SS, Woo WL,Parathai P

Discipline:Engineering

Volume:137

Page Number:124-129

Translation or Not:no