Speech Enhancement Algorithm Based On AD Neural Networks

Speech is one of the best natural and convenient intercommunication manners among humankind. It is also one of the ideal man-machine conversation interfaces. Nowadays, Speech processing technologies have been broadly used in many applied field. But speech under real environment is unavoidable to suffer the influence of environment all round. In order to minimize the effects of the noise on the performance of the processing systems, speech enhancement technology is applied in the various speech processing systems.Consequently the study of speech enhancement technology is very significant.This paper detail the speech enhancement technique in the circumstance of adding noise, such as spectral subtraction and LMS adaptive filtering. The paper summarizing the design feature and learning algorithm of neural, and introduces a method based on the combination of spectral subtraction and adaline linear neural networks in the enhancement of speech. First, enhance the speech with spectral subtraction to get a good envelope; then applie AD networks to the enhanced signal to have a high quality speech.MATLAB simulations indicates that the new algorithm can effectively restrain noise and increase signal-noise rate (SNR). At last we compare the performance of these methods in subjective and objective way. The experiment results prove that the method can greatly improve the quality and have better aural effect. The new algorithm also have good abilities of reducing the musical noise.