Please use this identifier to cite or link to this item:
Title: The application of support vector machine for speech classification
Authors: Gauci, Oliver
Debono, Carl James
Gatt, Edward
Micallef, Paul
Keywords: Support vector machines
Speech processing systems -- Computer programs
Machine learning
Classification -- Software
Issue Date: 2006
Publisher: University of Malta. Faculty of ICT
Citation: Gauci, O., Debono, C. J., Gatt, E., & Micallef, P. (2006). The application of support vector machine for speech classification. 4th Computer Science Annual Workshop (CSAW’06), Bighi. 1-4.
Abstract: For the classical statistical classification algorithms the probability distribution models are known. However, in many real life applications, such as speech recognition, there is not enough information about the probability distribution function. This is a very common scenario and poses a very serious restriction in classification. Support Vector Machines (SVMs) can help in such situations because they are distribution free algorithms that originated from statistical learning theory and Structural Risk Minimization (SRM). In the most basic approach SVMs use linearly separating Hyperplanes to create classification with maximal margins. However in application, the classification problem requires a constrained nonlinear approach to be taken during the learning stages, and a quadratic problem has to be solved. For the case where the classes cannot be linearly separable due to overlap, the SVM algorithm will transform the original input space into a higher dimensional feature space, where the new features are potentially linearly separable. In this paper we present a study on the performance of these classifiers when applied to speech classification and provide computational results on phonemes from the TIMIT database.
Appears in Collections:Scholarly Works - FacICTCCE
Scholarly Works - FacICTCS
Scholarly Works - FacICTMN

Files in This Item:
File Description SizeFormat 
Proceedings of CSAW'06 - A11.pdf88.34 kBAdobe PDFView/Open

Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.