Isolated digit recognition using a block diagonal recurrent neural network

Author

Sivakumar, S.C. ; Phillips, W.J. ; Robertson, W.

Author_Institution

Dalhousie Univ., Halifax, NS, Canada

Volume

2

fYear

2000

fDate

2000

Firstpage

726

Abstract

The objective of this paper is to recognize speech based on speech prediction techniques using a discrete time recurrent neural network (DTRNN) with a block diagonal feedback weight matrix called the block diagonal recurrent neural network (BDRNN). The ability of this network has been investigated for the TIMIT isolated digits spoken by a representative speaker. Simulation results for classifying the utterances show that the size of the BDRNN required is very small compared to multilayer perceptron networks with time delayed feedforward connections

Keywords

pattern classification; recurrent neural nets; speech recognition; TIMIT isolated digits; block diagonal feedback weight matrix; block diagonal recurrent neural network; discrete time recurrent neural network; isolated digit recognition; simulation results; speech prediction techniques; speech recognition; utterances classification; Character recognition; Computer architecture; Delay effects; Neurofeedback; Nonlinear dynamical systems; Output feedback; Recurrent neural networks; Speech recognition; State feedback; Symmetric matrices;

fLanguage

English

Publisher

ieee

Conference_Titel

Electrical and Computer Engineering, 2000 Canadian Conference on

Conference_Location

Halifax, NS

ISSN

0840-7789

Print_ISBN

0-7803-5957-7

Type

conf

DOI

10.1109/CCECE.2000.849560

Filename

849560