Itakura-Saito Divergence Non Negative Matrix Factorization with Application to Monaural Speech Separation

Abstract

Monaural source separation is an interesting area that has received much attention in the signal processing community as it is a pre-processing step in many applications. However, many solutions have been developed to achieve clean separation based on Non-Negative Matrix Factorization (NMF). In this work, we proposed a variant of Itakura-Saito Divergence NMF based on source filter model that captures the temporal continuity of speech signal. The algorithm shows a very good separation results for mixture of two speech sources in terms of artifacts reduction. Besides that, Source to distortion ratio (SDR) and Source to Artifact Ratio (SAR) were found to be higher when compared with NMF algorithms with Kullback-Leibler and Euclidean divergences.

Performance Evaluation of Selected Cost Functions in Non Negative Matrix Factorization Based Decomposition of Acoustic Mixture

“…Each of the speakers was made to make a sentence which varied in duration of roughly 4 to 8 seconds at a sampling frequency of 16kHz. The set up was similar to what is obtainable in (Adewusi et al, 2016). The magnitude spectrograms of the time-domain signal were obtained using the Short Time Fourier ( )…”

“…It usually comprises low-power transient components such as note tracks as well as higher power components such as tonal parts of sustained notes. IS has been successfully used for the improvement of the separation performance in monaural problem with group sparsity and temporal continuity (Lef`Evre et al, 2011;Fervotte et al, 2009;Adewusi et al, 2016).…”

Speech Separation Methodology for Hearing Aid

“…Fig. 2 represents the left and right input in the case of stereo audio signals that has been concatenated in time [34] where Fig. 2a shows the single dictionary of spectral atoms which is used to encode both channels via the two coefficient matrices H l dt and H r dt shown in Fig.…”