Phonemic speech recognition system based on a neural network

Author

Kepuska, Veton Z. ; Gowdy, John N.

Author_Institution

Dept. of Electr. & Comput. Eng., Clemson Univ., SC, USA

fYear

1989

fDate

9-12 Apr 1989

Firstpage

770

Abstract

A neural-network-based system for phonemic speech recognition is described. The self-organized feature map algorithm developed by T. Kohonen (1988) is used for the network model. The solution to the problem of overlap of feature vectors that correspond to speech phonemic units is proposed. This involves identification of transient segments of the speech, voiced vs. unvoiced determination, and phonemic context. The implementation takes form as a network with labels of transients and with up to two phonemic labels corresponding to stationary regions of the speech. The system is tested using a subset of an isolated-word database consisting of twenty words

Keywords

neural nets; speech recognition; feature vectors overlap; isolated-word database; network model; neural-network-based system; phonemic speech recognition; self-organized feature map algorithm; transient segments identification; Band pass filters; Bandwidth; Filter bank; Finite impulse response filter; IIR filters; Natural languages; Neural networks; Spectral analysis; Speech analysis; Speech recognition;

fLanguage

English

Publisher

ieee

Conference_Titel

Southeastcon '89. Proceedings. Energy and Information Technologies in the Southeast., IEEE

Conference_Location

Columbia, SC

Type

conf

DOI

10.1109/SECON.1989.132504

Filename

132504