The Speech Enhancement Based on the Speech Absence Probability and the Auditory Masking Effect

摘要

This paper proposes one speech enhancement method based on the speech absence probability and the auditory masking effect in this paper in order to improve the quality of the speech in the environment of the low SNR. This method dynamically determines the auditory masking threshold of the speech signal of each frame in all the key frequency bands by the way of the speech absence probability (SAP). Then we can adjust the subtraction parameters according to the auditory masking threshold adaptively, so as to achieve the purpose of removing the noise. The simulation results show that the method in this paper can more effectively reduce the musical noise than the other spectral subtraction. And it is more in line with the human auditory features. In particular, the speech has more clarity and intelligibility in the case of the low SNR.