Hai-yan YANG（杨海燕）, Xin-xing JING（景新幸）
College of Information and Communication, Guilin University of El ectronic and Technology, Guiln 541004，China
Abstract-In this paper, the specific application of key words spotting used in the network monitoring is studied, and the keywords spotting is emphasized. The whole monitoring system is divided into two modules: network mo nitoring and keywords spotting. In the part of network monitoring, this paper ad opts a method which is based on ARP spoofing technology to monitor the users’ d ata, and to obtain the original audio streams. In the part of keywords spotting, the extraction methods of PLP (one of the main characteristic parameters) is s tudied, and improved feature parameters-PMCC are put forward. Meanwhile, in or der to accurately detect syllable, the paper compares the double-threshold meth od with variance of frequency band method, and use the latter to carry out endpo int detection. Finally, keywords recognition module is built by HMM, and identif ication results are contrasted under Matlab environment. From the experiment res ults, a better solution for the application of key words recognition technology in network monitoring is found.
Key words-network monitoring; keywords spotting; PLP; P MCC; Hidden Markov Model(HMM)
Manuscript Number: 1674-8042(2011)02-0144-04
dio: 10.3969/j.issn.1674-8042.2011.02.11
References
[1]Xing-jun Yang, Hui-sheng Chi, 1995. Voice Digital Signal Processin g. Electronic Industry Press, Beijing, p.8.
[2]Bing-xi Wang, Dan Qu, 2005. Practical Fundamentals of Speech Recogn ition. National Defence Industry Press, Beijing, p.287.
[3]Xiang Zhang, Jiang-wei Zhang, 2006. An Interpretation of ARP Cheat. 2006 Graduate of Beijing University Academic Exchange-Conference on Communicat ions and Information Technology, p.1776.
[4]Bing Wu, Ting-gen Shen, Xue-hua Song, Sheng-hua Xu, 2008. The MFC C Feature Extraction Based on The Noisy Situation. Microcomputer Infor mation(Control Automation), 24.1-1: 224-226.
[5]Ying-chun Xie, Xiang-zhen Yu, Jiang-ping Liu, Wei-hua Zhang, 200 5. Speaker Identification Based on Efficiently Combining Manifold Features. Modern Electronic Technology, 9: 68-70.
[6]L. R. Rabiner, 1979. Speaker-Independent Recognition of Isolated Wo rds Using Clustering Techniques. IEEE on ASSP.
[full text view]
