BLIND SEPARATION OF ACOUSTIC MIXTURES BASED ON LINEAR PREDICTION ANALYSIS
Kostas Kokkinakis, Vicente Zarzoso, Asoke Kumar Nandi · 2003
In this paper we propose a general method for separating mixtures of multiple audio signals observed in a real acoustic environment. The multipath nature of acoustic propagation is addressed by the use of the FIR polynomial matrix algebra, while spatio-temporal separation is achieved by entropy maximization using the natural gradient algorithm. The undesired temporal whiteness of the es-timates is overcome with the use of linear prediction (LP) analysis. As opposed to a previous LP-based method, no assumptions on re-lative strengths of individual sources to specific mixtures are made. Other benefits such as reduced computational complexity and in-creased convergence speed are also emphasized. Finally, a number of experiments demonstrate the validity and general applicability of the proposed method. 1.