In this paper, we proposed applying Hadamard Error-Correcting Output Code to extend binary classifier to multi-class problems. Compared with other ECOC approaches, Hadamard ECOC is easy to construct and suitable to any number of classes. We combine it with binary support vector machine (SVM) to solve the multi-class problem of speaker identification. Compared with the traditional'1-against-rest'method, the experiment result shows that Hadamard ECOC has much better and more stable performance to any number of classes for the multi-class problem and is robust with respect to the assignment of distributed representations to particular classes.