Title :
Keyword spotting method based on speech feature space trace matching
Author :
Wu, Ya-Dong ; Liu, Bao-Long
Author_Institution :
Dept. of Comput. Sci. & Eng., Shanghai Jiao Tong Univ., China
Abstract :
Keyword spotting (KWS) has been an active research area in recent years. Based on the theory of feature space trace time normalization, an efficient keyword spotting method is proposed in this paper, with the effect of the fragment trace length probed. Experiments show that the performance of this method is close to that of manual spotting, having some practicability.
Keywords :
speaker recognition; speech processing; KWS method; feature space trace time normalization; fragment trace length; keyword spotting method; manual spotting; speech feature space trace matching; Background noise; Hidden Markov models; Monitoring; Neural networks; Signal processing; Speech enhancement; Speech processing; Speech recognition; Testing; Vocabulary;
Conference_Titel :
Machine Learning and Cybernetics, 2003 International Conference on
Print_ISBN :
0-7803-8131-9
DOI :
10.1109/ICMLC.2003.1260128
