Phoneme sequence pattern recognition using fuzzy neural network

Author

Kwan, H.K. ; Dong, X.

Author_Institution

Dept. of Electr. & Comput. Eng., Windsor Univ., Ont., Canada

Volume

1

fYear

2003

fDate

14-17 Dec. 2003

Firstpage

535

Abstract

In this paper, a 2-D phoneme sequence pattern recognition using the fuzzy neural network is presented. The self-organizing map and the learning vector quantization are used to organize the phoneme feature vectors of short and long phonemes segmented from speech samples to obtain the phoneme maps. The 2-D phoneme response sequences of the speech samples are formed on the phoneme maps by the Viterbi search algorithm. These 2-D phoneme response sequence curves are used as inputs to the fuzzy neural network for training and recognition of 0-9 digit-voice utterances. Simulation results are given.

Keywords

fuzzy neural nets; maximum likelihood estimation; self-organising feature maps; speech recognition; Viterbi search algorithm; fuzzy neural network; learning vector quantization; pattern recognition; phoneme sequence; self-organizing map; Artificial neural networks; Feature extraction; Fuzzy neural networks; Hidden Markov models; Mel frequency cepstral coefficient; Neurons; Organizing; Pattern recognition; Speech recognition; Vector quantization;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks and Signal Processing, 2003. Proceedings of the 2003 International Conference on

Conference_Location

Nanjing

Print_ISBN

0-7803-7702-8

Type

conf

DOI

10.1109/ICNNSP.2003.1279329

Filename

1279329