Title :
Support Vector Machines with the priorities method for speaker independent 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 speaker independent phoneme recognition system, based on Support Vector Machines (SVMs) method was improved by adding a priority scheme to forecast the three most likely phonemes. The system helps improve the obtained recognitions rate . For the phoneme recognition system, four multiclass SVMs methods, the All- at-once, One-against-all, One-against-one, and the Directed Acyclic Graph SVM (DAGSVM), were designed. The One-against-one method performed best, achieving an accuracy of 53.70%. This accuracy was further increased to 75.41%, when the second and third priorities were considered in the priorities method. All tests were carried out on the TIMIT database.
Keywords :
directed graphs; speaker recognition; support vector machines; TIMIT database; all-at-once method; directed acyclic graph SVM; one-against-all method; one-against-one method; priority method; speaker independent phoneme recognition; support vector machine; Accuracy; Discrete wavelet transforms; Kernel; Speech; Speech recognition; Support vector machines; Training; multiclass support vector machines; phoneme recognition; priorities method; speaker-independent;
Conference_Titel :
Signal Processing and Information Technology (ISSPIT), 2011 IEEE International Symposium on
Conference_Location :
Bilbao
Print_ISBN :
978-1-4673-0752-9
Electronic_ISBN :
978-1-4673-0751-2
DOI :
10.1109/ISSPIT.2011.6151597
