Blind Speech Separation Based on Homotopy Non-linear Model

Blind source extraction (BSE) algorithm based on linear prediction model can be used to separate speech signals. The best advantage of this algorithm is that it is very simple and easy to compute. This algorithm is based on the fact that speech signals can be modeled by linear prediction process. But homotopy non-linear model is better than linear prediction model for speech signals. A new BSE algorithm based on homotopy non-linear model is proposed to separate speech signals. Computer simulations show that the proposed algorithm has better performance than the BSE algorithm based on linear prediction model.