A improved dual-threshold speech endpoint detection algorithm
Qiuyu Guo, Nan Li, Guangrong Ji · 2010
This paper presents an improved dual-threshold speech endpoint detection algorithm based on short-time magnitude increment and short-time zero-crossing rate, which uses the continuous rise of speech short-time magnitude as the first-level decision threshold and short-time zero-crossing rate as the second-level decision threshold. Experimental results show that this algorithm is accurate, easy and reliable in the case of ideal signal to noise ratio(SNR).