Application In Speech Enhancement Of Mathematics Morphology

The system of speech recognition is to work in the environment of peace mostly .In the environment of noise especially strong noise, the recognition rate of the system of speech recognition will get serious influenced. The major goal of speech enhancement is extraction as far as possible pure original speech sound from the noisy signal, and speech enhancement is used in speech recognition as a kind of very effective method.Digital morphological filter is a branch of nonlinear filter field, which is the most important and quickly development. In recent years, it has gotten successful application in image processing, biological medical science and such aspect. Because speech signal is a kind of complex, non-stationary and nonlinear signal as image signal, and their digital processing is similar, it is possible morphology can be also used in the research field of the signal of speech .This paper works mainly in the application algorithm of morphological speech enhancement technique. A serial generalized morphological filter with multi-structural element is used suppression white Gaussian noise or pulse noise embedded in the speech signal. The paper compares morphological speech enhancement algorithm with classical approach on the feature of speech in the frequency domain and time domain. Morphological filter is a kind of new nonlinear speech enhancement algorithm. Simulation experiment results of Matlab confirm this kind of the nonlinear processing technique of speech is feasible. Moreover in speech enhancement, especially in reducing the pulse noise, morphological algorithm has its unique advantage. Particularly morphological filter may maintain the preferable accurate of the speech signal in speech waveform, and which produces little impairment to the formant of speech .So the spectrum structure of the speech is retained well, and the quality of the speech will not be reduced. Because of the quickness in calculating and briefness in algorithm, MM will be a preferable approach to nonlinear speech signal processing.