Research on speech separation based on independent component analysis and wavelet filtering

NIE Xiao-yan

DOI: https://doi.org/10.3969/j.issn.1004-373X.2012.19.019

2012-01-01

Abstract:Separation of the mixed speech signals is the important content of the blind source separation,and a difficult problem in the fields of signal processing.Firstly,the original signal was denoised using wavelet filtering.And then the independent components of the mixed signal were separated by Infomax ICA algorithm.Experiment results show that this algorithm can not only eliminate noise compared with the traditional filter method,but also the details of the other signal almost have no damage.This algorithm can fully separate the speech signals with same or similar frequency,and denoising performance of the algorithm is better than that of the traditional filter method.