Title :
A fast algorithm for large vocabulary keyword spotting application
Author :
Huang, Eng-Fong ; Wang, Hsiao-Chum ; Soong, Frank K.
Author_Institution :
Telecommun. Lab., Minist. of Commun., Chung-Li, Taiwan
fDate :
7/1/1994 12:00:00 AM
Abstract :
Presents a fast algorithm for spotting a large number of keywords in unconstrained, continuous speech using an HMM-based continuous speech recognizer. This fast algorithm is based on a two stage scheme. In the first stage, the forward backward search is performed for detecting N most likely common subwords. In the second stage, the tree-trellis search is carried out to determine the optimum keyword by traversing the tree-structural vocabulary in an effective way. Compared with the conventional whole-word based keyword spotting algorithm, the proposed fast algorithm can drastically reduce the computational cost
Keywords :
filtering and prediction theory; hidden Markov models; speech recognition; trees (mathematics); HMM; computational cost reduction; continuous speech recognizer; fast algorithm; ffilter models; forward backward search; large vocabulary keyword spotting; tree-structural vocabulary; tree-trellis search; unconstrained continuous speech; Computational efficiency; Health information management; Hidden Markov models; Speech recognition; Viterbi algorithm; Vocabulary;
Journal_Title :
Speech and Audio Processing, IEEE Transactions on
