Microphone Array Speech Enhancement In Noisy Environments, Technology Research

In the voice processing system, the microphone received signal is often subjected to interference of the external environment, which makes the microphone receives the voice signal often with a noise signal. Especially in a noisy environment, noise and interference seriously affect the intelligibility of speech and the voice system processing performance, such as the hum of the machines in the plant, aircraft cockpit engine noise emitted, as well as bar, KTV etc, the noise will hamper the communication between people, affecting the health of the human body. Therefore, people have been seeking effective noise suppression methods to enhance the quality of the voice. In the general noise environment, the use of single-microphone speech enhancement method can suppress noise well, but the performance of noise suppression will degrade in noise environment. People have been proposed microphone array speech enhancement method to suppress noise, microphone array have spatial selectivity and high signal gain. In recent years, microphone array become a popular technology.In this article, we described speech enhancement based microphone array, microphone array capture voice signal, then we use beamforming and interference microphone cancellation to enhance the desired speech signal and suppress interfering signals in noisy environment. The main work of this paper is to compare several time delay estimation method, and select the algorithm for noise environment in this article; this article combines voice activity detection algorithms with time delay estimation, in the condition of the false alarm probability is small or zero to reduce the probability of miss alarm, to further improve the performance of speech enhancement algorithm. According to the results of voice activity detection and delay estimation, we determine the current frame signal is the desired speech signal, interference or other signal. If the current signal is desired speech signal, we adaptively enhance the desired signal, then the interference cancellation further eliminate noise; interfering signals, the array signal adaptively cancel the interference signal; if the other signal, we enhance the voice,then elimate the interference and noise. The simulation results show that the speech enhancement algorithm of this article has better noise suppression in noisy environments, enhance the effect of the voice.