Study Of Speaker Recognition System Based On Support Vector Machines

Support Vector Machine (SVM) is an important learning method of statistical learning theory,and is a powerful tool for pattern recognition problems.This paper studies the speaker recognition problem using support vector machine,and persents a SVM method on little-scale training data.By analyzing speech characteristic parameters and the basis methods of speaker recognition,we choose MFCC and its difference to be the characteristic parameters.Two text-dependent speaker identification systems based on DTW and SVM were implemented. Finally, a real speech library was built.The above mentioned methods are simulated and the result of simulation is given.