Please use this identifier to cite or link to this item:
Title: Discrete wavelet transforms with multiclass SVM for phoneme recognition
Authors: Cutajar, Michelle
Gatt, Edward
Grech, Ivan
Casha, Owen
Micallef, Joseph
Keywords: Support vector machines
Speech processing systems
Radial basis functions
Wavelets (Mathematics)
Kernel functions
Issue Date: 2013
Publisher: Institute of Electrical and Electronics Engineers Inc.
Citation: Cutajar, M., Gatt, E., Grech, I., Casha, O., & Micallef, J. (2013). Discrete wavelet transforms with multiclass SVM for phoneme recognition. EuroCon 2013, Zagreb. 1695-1700.
Abstract: A phoneme recognition system based on Discrete Wavelet Transforms (DWT) and Support Vector Machines (SVMs), is designed for multi-speaker continuous speech environments. Phonemes are divided into frames, and the DWTs are adopted, to obtain fixed dimensional feature vectors. For the multiclass SVM, the One-against-one method with the RBF kernel was implemented. To further improve the accuracies obtained, a priority scheme was added, to forecast the three most likely phonemes. After classification, all frames were again re-grouped, in order to evaluate the accuracy of the system according to the substitution, deletion and insertion errors. The percentage correct and accuracy, obtained from the designed phoneme recognition system, were 63.08% and 53.27% respectively. All tests were carried out on the TIMIT database. This phoneme recognition system is intended to be implemented on a dedicated chip, to improve the speed of the software implementation by approximately 100 times.
Appears in Collections:Scholarly Works - FacICTMN

Files in This Item:
File Description SizeFormat 
Discrete Wavelet Transforms with Multiclass SVM for Phoneme Recognition.pdf
  Restricted Access
Discrete wavelet transforms with multiclass SVM for phoneme recognition471.42 kBAdobe PDFView/Open Request a copy

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