Speech Recognition Oriented Vowel Classification Using Temporal Radial Basis Functions

TL;DR

This paper introduces a temporal radial basis function approach for vowel classification in speech recognition, achieving high accuracy and demonstrating advantages like speed and invariance.

Contribution

It presents a novel TRBF-based method for vowel classification that outperforms traditional techniques in accuracy and efficiency.

Findings

98.06% training accuracy

90.13% test accuracy

Effective for natural speech vowels

Abstract

The recent resurgence of interest in spatio-temporal neural network as speech recognition tool motivates the present investigation. In this paper an approach was developed based on temporal radial basis function "TRBF" looking to many advantages: few parameters, speed convergence and time invariance. This application aims to identify vowels taken from natural speech samples from the Timit corpus of American speech. We report a recognition accuracy of 98.06 percent in training and 90.13 in test on a subset of 6 vowel phonemes, with the possibility to expend the vowel sets in future.