Voice Activity Detection in Complex Environment Based on Hilbert-Huang Transform and Order Statistics Filter

Hilbert-Huang Transform (HHT) is a fully data driven adaptive non-stationary signal time-frequency analysis method. But the Hilbert energy spectrum curve of speech signal is fluctuate in strong noise environment，it has a great influence to voice activity detection. So an effective voice activity detection algorithm is proposed based on HHT and Order Statistics Filter (OSF) in this paper. This method first decompose noise signal into intrinsic mode functions by empirical mode decomposition. Then the Hilbert energy spectrum is synthesized by adaptive weight selection of each intrinsic mode functions, through OSF to smooth the energy spectrum. Finally, the speech and noise divergence is judged by means of the smoothed energy spectrum. Experimental results show obviously that under complex noisy environment, this method is still able to effectively detect the speech signal, and reduce the error detection rate in low signal to noise ratio conditions.