Joint time-frequency analysis for speech recognition

Authors

Albert James M. Licup Department of Physics, University of San Carlos
Edcel John L. Salumbides Department of Physics, University of San Carlos
Bernardino J. Buenaobra Department of Physics, University of San Carlos
Kees Karremans Vrije Universiteit Amsterdam, The Netherlands and University of San Carlos, Philippines

Abstract

In this paper we present a system developed to recognize a set of words. The feature vectors used are derived from joint time-frequency analysis of the speech signal. We used a multilayer feed-forward neural network trained with the error back propagation algorithm as the feature classifier. We have investigated the network structure that gives the optimum performance. Using a neural network with two hidden layers the system was able to recognize twenty words with 98% recognition rate.

Downloads

Issue

2001: Proceedings of the 19th Samahang Pisika ng Pilipinas Philippine Physics Congress

Article ID

SPP-2001-PP-10

Section

Image and Signal Processing

Published

2001-10-24

How to Cite

[1]

AJM Licup, EJL Salumbides, BJ Buenaobra, and K Karremans, Joint time-frequency analysis for speech recognition, Proceedings of the Samahang Pisika ng Pilipinas 19, SPP-2001-PP-10 (2001). URL: https://proceedings.spp-online.org/article/view/SPP-2001-PP-10.