Title :
A novel Chinese continuous speech endpoint detection method based on time domain features of the word structure
Author :
He, SuNing ; Yu, Juebang
Author_Institution :
Coll. of Electron. Eng., Univ. of Electron. Sci. & Technol. of China, Chengdu, China
fDate :
29 June-1 July 2002
Abstract :
Based on the time domain features of Chinese words, which are short time peak-valley energy and zero crossing, we propose a novel method for continuous speech endpoint detection. It is simple and easy to use, with high detection rate and low computational complexity. The effectiveness of the method has been verified by experiments performed on some continuous Chinese speech. Experimental results show that 96% successful endpoint detection rate can be reached for "863" speech. We have also found that the endpoint detection rate could be enhanced further if we take measures to overcome the effect of some inherent speech factors such as the speaker\´s speaking style, speed, coarticulation, stressing or muting.
Keywords :
computational complexity; feature extraction; speech recognition; time-domain analysis; Chinese continuous speech endpoint detection; coarticulation; computational complexity; peak-valley energy; speaking speed; speaking style; speech muting; speech stressing; time domain features; word structure; zero crossing; Design methodology; Educational institutions; Gas detectors; Helium; Mouth; Phase detection; Power engineering and energy; Speech enhancement; Speech recognition; Velocity measurement;
Conference_Titel :
Communications, Circuits and Systems and West Sino Expositions, IEEE 2002 International Conference on
Print_ISBN :
0-7803-7547-5
DOI :
10.1109/ICCCAS.2002.1178954
