Support Vector Machines is a new and promising technique in statistical learning theory. Recently, this technique produced very interesting results in pattern recognition. In this paper, one of the first application of Support Vector Machines(SVM) technique for the problem of keyword spotting is presented. A two-class SVM approach is first proposed which classifies the correct and the incorrect keywords. The second method uses multi-class SVM, it assigns to each keyword a class label This is a first work proposed to use two-class SVM and multi-class SVin keyword spotting, in order to improve recognition accuracy. The results obtained are very promising.