Single Channel Speech Blind Separation Based On Genetic Algorithm Optimization

Blind signal separation (BSS) technology is a new research direction in the field of modern signal processing. In this paper, a single channel speech blind separation method based on time-frequency masking and genetic algorithm optimization is proposed for single-channel speech blind separation. Firstly, the mixed signal is decomposed into an Intrinsic Mode Function (IMF) with different source signal characteristics by using the Ensemble Empirical Mode Decomposition (EEMD) algorithm to compose a new multidimensional signal, and then use the genetic algorithm based on genetic algorithm Optimization of Independent Component Analysis Method to Realize Blind Separation of Signals. The experimental results show that the method can effectively improve the efficiency and stability of the operation and obtain a good separation effect.