Adaptive Sample Entropy of Speech Endpoint Detection
Jing Xinxin · Audio Engineering · 2014
A new speech endpoint detection algorithm based on adaptive sample entropy is proposed. The sample entropy combine with zero crossing ratemethodin is used in order to improve the detection accuracy,good detection effect is obtained through an adjustable parameters of the threshold H. Simulation in MATLAB,the results show that when the signal-to-noise ratio dropped to zero,the method based on adaptive sample entropy detection accuracy can still reach 90%,the detection performance fully meet the design requirements,and is superior to the traditional method based on energy.