Speech Enhancement Based on Double RBF Networks

In this paper, a non-linear spectral estimation for noise reduction is present which is approximated and implemented by double Radial Basis Function (RBF) networks. The simulation results indicate that the method can greatly improve the quality and the intelligibility of speech, and have other advantages such as the widely applicable Signal-to-Noise Ratio (SNR) range, less computation load. Particularly the method may maintain the preferable accurate of signal in speech waveform, and the quality of speech signals have been improved obviously.