The Research On The Voice Activity Detection In The Environment With High Noise

Generally, in many application of speech signal processing, the quality of voice has declined because of background noise.In recent years, with the development of application area in voice communication, people take more focus on voice communication in background with high noise, especially about voice communication in tank and armored car.Due to this reason, how to successfully and precisely detect the presence of voice, how to exact as much original voice data from the corrupted signal and introduce as little distortion as possible has been an more and more important research area.Through the analysis of noise signal and speech signal, a new method for voice activity detect is presented in this paper. The method utilizes the character of short-time energy, can adjust threshold according to the movement of the noise. The circuit takes Motorola's DSPF801 chip as main control unit.Experiments show this method has high performance even in a low SNR circumstance. Besides, it has the advantage of simple computing and easy hardware processing. Its high reliability makes it suitable for application in tank and armored car.Finally ,we suggest some improved schemes and the prospect about the project.