Application of Vowel Recognition Model Based on Improved SVM Algorithm

A vowel is a sound where air coming from the lungs is not blocked by the mouth or throat. The articulatory features that distinguish different vowel sounds are said to determine the vowel's quality. It is very important to recognize different vowel classes. A vowel recognition model based on improved least squares support vector machine (LSSVM) is presented. In the model, when the LSSVM is used in vowel recognition, it is presented to choose parameter of kernel function on dynamic, which enhances preciseness rate of diagnosis. The Fibonacci symmetry searching algorithm is simplified and improved. The changing rule of kernel function searching region and best shortening step is studied. The best pattern recognition results are obtained by means of synthesizing kernel function searching region and best shortening step. The simulation results show the validity of the vowel recognition model.

[1]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[2]  Sarah Hawkins,et al.  Formant frequencies of RP monophthongs in four age groups of speakers , 2005, Journal of the International Phonetic Association.

[3]  Guodong Guo,et al.  Support vector machines for face recognition , 2001, Image Vis. Comput..

[4]  Alexander J. Smola,et al.  Support Vector Method for Function Approximation, Regression Estimation and Signal Processing , 1996, NIPS.

[5]  Michel Verleysen,et al.  Fast bootstrap for least-square support vector machines , 2004, ESANN.

[6]  Chih-Jen Lin,et al.  A comparison of methods for multiclass support vector machines , 2002, IEEE Trans. Neural Networks.

[7]  Shie Mannor,et al.  The kernel recursive least-squares algorithm , 2004, IEEE Transactions on Signal Processing.

[8]  Matthew Gordon,et al.  Phonetic structures of Turkish Kabardian , 2006, Journal of the International Phonetic Association.