Blind Separation of Two Human Speech Signals using Natural Gradient Algorithm by Employing the Assumptions of Independent Component Analysis

Abstract

This paper uses the Natural Gradient Algorithm (NGA) to separate two mixed signals into their original components using the ICA assumption of statistical independence of the source signals. The NGA used is formulated using instantaneous Blind Signal Processing where time delay is not factored in the computation of the independent signals. The design uses a 2 x 2 Multiple Input Multiple Output (MIMO) system to accept the two blind speech signals, mix them and separate them to retain their original form or their filtered version. The Fibonacci activation function is used in iterating the coefficients of the NGA up to 1000 times where the best separation is achieve. The result shows that the separation is achieved to the level where the wave formation and frequency spectrum significantly reflect those of the input signals and on listening to the separated signal resembles the original signals, as well.