Title :
An Improved Mandarin Keyword Spotting System Using MCE Training and Context-Enhanced Verification
Author :
Liang, JiaEn ; Meng, Meng ; Wang, XiaoRui ; Ding, Peng ; Xu, Bo
Author_Institution :
Inst. of Autom., Chinese Acad. of Sci., Beijing
Abstract :
The task of keyword spotting is to detect a set of keywords in the input continuous speech. The main goal of this work is to develop an improved Mandarin keyword spotting (KWS) system for conversational telephone speech (CTS). In this paper, we propose an efficient online-garbage model based KWS system, which integrated with a word-level minimum classification error (MCE) training method and a novel context-enhanced verification method. Experiment showed that the proposed methods can reduce the equal-error-rate (EER) of the system by 13.8% in relative
Keywords :
natural languages; speech processing; speech recognition; Mandarin keyword spotting system; context-enhanced verification; context-enhanced verification method; conversational telephone speech; equal-error-rate; online-garbage model; word-level minimum classification error; Acoustic beams; Automation; Context modeling; Contracts; Filters; Hidden Markov models; Speech processing; Telephony; Viterbi algorithm; Vocabulary;
Conference_Titel :
Acoustics, Speech and Signal Processing, 2006. ICASSP 2006 Proceedings. 2006 IEEE International Conference on
Conference_Location :
Toulouse
Print_ISBN :
1-4244-0469-X
DOI :
10.1109/ICASSP.2006.1660228
