Electrical and Computer Engineering Department Main MenuBradley University Main Web Site for Prospective Students for Current Students for Our Alumni for Faculty Contact Information Faculty Directory Senior Projects Labs and Research Areas Useful links ECE Department Home Page Senior Projects


Speech Recognition Using a Neural Network
Andrey Pilipchak
Advisors: Dr. T. Stewart and Dr. I. S. Ahn

In this paper, a neural network architecture is presented for speech recognition. A library of two words is created by capturing male and female 8-bit voice data using a SoundBlaster sound card on an IBM-PC. The captured data are sent through four different band-pass filters for feature extraction of each word presented. Then the band-pass filter outputs are put through two cascaded low-pass filters resulting in down-sampling. The low-pass filter outputs are presented to an error backpropagation neural network for training. MATLAB, a mathematical software environment, is used for the training and testing of the algorithm. The trained neural network performs reliably distinguishing the two words regardless of speaker and gender.

[an error occurred while processing this directive]
[Prospective Students] [Current Students] [Alumni] [Faculty]
[Home] [Contact us] [Curriculum] [Senior Projects] [Research] [People] [Links]
Copyright (c)1995-2013 Bradley University. All rights reserved.
. .