A Neural Network Based Effective Quality Speech Enhancement Using Monte Carlo Method

Degrading the quality and intelligibility of speech signals, background noise plays a severe problem in communication and other speech related systems. In order to get rid of this problem, it is important to enhance the original speech signal mainly through noise reduction. Since speech signals are non-linear and non-stationary in nature, the performance of related studies is significantly dependent on the analysis method. Although Fourier transform and wavelet analysis made great contributions they suffer from many shortcomings in the case of nonlinear and non-stationary signals. Particle filters (PF) provide solution to this problem since it allows the noise to be non-stationary and non-linear. It also decreases the computational cost by allowing us to use less particles. The simulation is done using MATLAB and the performance of PF for various speech signals was computed.