Title :
Discrete wavelet transforms with multiclass SVM for phoneme recognition
Author :
Cutajar, M. ; Gatt, E. ; Grech, I. ; Casha, O. ; Micallef, J.
Author_Institution :
Dept. of Microelectron. & Nanoelectron., Univ. of Malta, Msida, Malta
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.
Keywords :
audio databases; discrete wavelet transforms; radial basis function networks; speech processing; DWT; RBF kernel; TIMIT database; discrete wavelet transforms; fixed dimensional feature vectors; multiclass SVM; multispeaker continuous speech environments; phoneme recognition system; radial basis function; software implementation; support vector machines; Accuracy; Discrete wavelet transforms; Kernel; Speech; Speech recognition; Support vector machine classification; discrete wavelet transforms; phoneme recognition; speaker-independent; support vector machines;
Conference_Titel :
EUROCON, 2013 IEEE
Conference_Location :
Zagreb
Print_ISBN :
978-1-4673-2230-0
DOI :
10.1109/EUROCON.2013.6625205
