An improved SVM using predator prey optimization and Hooke-Jeeves method for speech recognition

Abstract

For automatic speech recognition, speech signal is represented in terms of featureset. Size of the feature set is an important aspect as it affects recognition accuracy,computational time, memory requirement and complexity of the model. Supportvector machine has good application prospects for speech recognition; nevertheless,performance of support vector machine is affected by its parameters. In this researchwork, a hybrid optimization technique is proposed to improve the learning abilityof support vector machine and to select the most appropriate feature set. The hybridtechnique integrates predator-prey optimization and Hooke-Jeeves method. To dealwith mixed type of decision variables, binary predator-prey optimization technique hasalso been introduced. During the initial phase, search is performed by predator-preyoptimization and to further exploit the search, Hooke-Jeeves method is applied. Theproposed technique with support vector machine has been implemented to recognizeTI-46 isolated word database in clean as well as noisy conditions and self-recordedHindi numeral database. The experimental results obtained by proposed techniquewith support vector machine shows improved recognition rate. Furthermore ROCcurve is analysed to verify sensitivity and specificity of results obtained by proposedtechnique with support vector machine.