Voice activity detection algorithm based on entropy in noisy environment
Cao Yali, Ying Hua Guan, La Dongsheng, Gao Jing · 2016
This paper proposes a novel voice activity detection algorithm to improve the speech detection performance in noisy environments. In the proposed algorithm, Natural weighted factor is introduced to improve the performance of speech signal detection based on wavelet energy entropy. We have evaluated system performance under noisy environments using whispered speech and noise. Experimental results indicate that our approach performs well in the degraded environment.