DocumentCode
1739531
Title
A hybrid model of hidden Markov models and a self-organizing neural network model in speech recognition
Author
Jingjiao, Li ; Jie, Sun ; Yanquan, Li
Author_Institution
Sch. of Inf. Sci. & Eng., Northeastern Univ., Shenyang, China
Volume
2
fYear
2000
fDate
2000
Firstpage
742
Abstract
This paper proposed a hybrid model based on hidden Markov models and a self-organizing neural network model, which can be used to speech recognition. The algorithm of training and adjustment coefficient vectors is given. Experimental results demonstrate the efficiency of the new algorithm in speech recognition
Keywords
feature extraction; hidden Markov models; learning (artificial intelligence); neural nets; self-organising feature maps; speech recognition; HMM; adjustment coefficient vectors; algorithm efficiency; feature extraction; hidden Markov models; hybrid model; self-organizing neural network model; speech recognition; training algorithm; Hidden Markov models; Information science; Intelligent networks; Iterative algorithms; Maximum likelihood estimation; Neural networks; Organizing; Speech recognition; Sun; Viterbi algorithm;
fLanguage
English
Publisher
ieee
Conference_Titel
Signal Processing Proceedings, 2000. WCCC-ICSP 2000. 5th International Conference on
Conference_Location
Beijing
Print_ISBN
0-7803-5747-7
Type
conf
DOI
10.1109/ICOSP.2000.891618
Filename
891618
Link To Document