Study Of Speech Enhancement Algorithm

In practical applications, such as speech telecommunication, speech synthesis, speaker recognition, the degradation of speech by all kinds of noise is inevitable and it affects the performance of the speech equipments very badly. So, it is necessary to denoise for the degraded speech, this is called SPEECH ENHANCEMENT.The approaches of speech enhancement are numerous. The first part of this paper is the origin and history of speech enhancement, several enhancement approaches and some knowledge related to speech enhancement are given too, for example, the models of speech and noise, and the estimation of noise. Then, speech enhancement based on wavelet threshold denoising follows, simulation result is provided. Meanwhile, some modification is given to hard threshold approach for its drawback, simulation result shows performance of the modified approach is better than the original one. But realization of the wavelet denoising with DSP is complex and its real-time is bad, so this paper discusses MMSE speech enhancement based on masking properties in detail: the reason of MMSE speech enhancement's bad real-time is given, at the same time, the modification is present and it improves MMSE speech enhancement's real-time, lastly, MMSE speech enhancement incorporating masking properties is introduced, it can keep more components of speech signal, this decreases speech distortion to some extent, evaluation result shows: ways mentioned above is workable and they are satisfactory.