DocumentCode
314026
Title
Cantonese phonemes recognition using modular neural network approach
Author
Hui, W.M. ; Liu, P.C.K. ; Li, K.C.
Author_Institution
Dept. of Electron. Eng., Hong Kong Polytech., Kowloon, Hong Kong
fYear
1997
fDate
29 Jun-4 Jul 1997
Firstpage
210
Abstract
The authors describe the design of a Cantonese speech recognition system based on modular neural networks to recognize all the Cantonese phonemes. The results showed a correct recognition of 98% in the database training set and over 85% in the testing set
Keywords
delays; learning (artificial intelligence); modules; multilayer perceptrons; natural languages; speech processing; speech recognition; Cantonese phonemes recognition; Cantonese speech recognition system; correct recognition; database; modular neural network approach; speech analysis; testing set; three layer MLP neural network; time-delay neural network; training set; Artificial neural networks; Databases; Large-scale systems; Neural networks; Neurons; Speech recognition; Speech synthesis; Telephony; Testing; User interfaces;
fLanguage
English
Publisher
ieee
Conference_Titel
Information Theory. 1997. Proceedings., 1997 IEEE International Symposium on
Conference_Location
Ulm
Print_ISBN
0-7803-3956-8
Type
conf
DOI
10.1109/ISIT.1997.613125
Filename
613125
Link To Document