|
|
|
| Voice Activity Detection Using Parallel Distance of Probability Density |
| Wu Qi-qian Zhang Xiong-wei Zou Xia |
| Department of Electronic Information Engineering, ICE PLAUST, Nanjing 210007, China |
|
|
|
|
Abstract In this paper, a Voice Activity Detection algorithm is developed, which uses a distance measure parameter – parallel connection distance of probability density to improve detection. The algorithm is based on the determination of the speech/noise divergence by means of parallel distance of the Mel-field sub-band log-energies probability density. The proposed VAD is tested on speech signals with various noises under different SNRs. Experimental results show that the proposed VAD outperforms the standard algorithm G.729 VAD.
|
|
Received: 28 May 2007
|
|
|
|
Corresponding Authors:
Wu Qi-qian
|
|
|
|
|
|
|
2008, Vol. 30 
