Title :
Sensitive keyword spotting for voice alarm systems
Author :
Chunlei Zhu ; Qing-Jie Kong ; Zhou, Liang ; Gang Xiong ; Fenghua Zhu
Author_Institution :
Dongguan Res. Inst., Cloud Comput. Center, CASIA, Dongguan, China
Abstract :
Keyword spotting is the task of identifying the occurrences of certain desired keywords in an arbitrary speech signal. Keyword spotting has many applications. One of them is emergency voice alarm systems, like command control and smart monitoring ATM etc. This paper presents a sensitive keyword spotting system applied to voice alarm occasions. By sensitive keyword retrieval, emergency warning systems can automatically monitor and follow up the situation, find the crisis occurs, and then give the alarm. Automatically retrieving and tracking for the particular occasion can not only save a lot of manpower and material resources, but is also more efficient. So the study of this subject has great reality and application value.
Keywords :
alarm systems; emergency services; information retrieval; speech recognition; arbitrary speech signal; emergency voice alarm systems; emergency warning systems; sensitive keyword retrieval; sensitive keyword spotting system; Acoustics; Alarm systems; Feature extraction; Hidden Markov models; Speech; Speech recognition; Viterbi algorithm; automatic tracking; keyword spotting; smart monitoring; voice alarm systems;
Conference_Titel :
Service Operations and Logistics, and Informatics (SOLI), 2013 IEEE International Conference on
Conference_Location :
Dongguan
Print_ISBN :
978-1-4799-0529-4
DOI :
10.1109/SOLI.2013.6611439