Research Of Multiclass Classification Support Vector Machine In Embedded Speech Recognition System

Speech recognition is an important aspect of speech signal processing. It is the foundation of human-computer interaction technology and has wide application prospect. It has great theoretical value and practical significance for us to do research on speech recognition.Automatic Speech Recognition is essentially a problem of pattern multiclass classification, Support Vector Machine(SVM) classifiers could provide an appropriate solution, since they are very well adapted to high-dimensional classification problems. The approach is systematic and motivated by statistical learining theory and Bayesian arguments. The training task involves optimization of a convex cost function:there are no false local minima to complicate the learning process. The approach has many other benefits, for example, the model constructed has an explicit dependence on the most informative patterns in the data(the support vectors), hence interpretation is straightforward. This paper constructed three non-specific person words speech recognition systems which are based on multiclass classification support vector machines of1vR,1v1, and DAG methods and we analysis and compare the performance of three classifier algorithms with current speech recognition method, then did a lot of simulation experiments. The experimental results show that the recognition results of speech recognition systems which are based on three multiclass classification algorithms are very good and better than the recognition results that is based on hidden markov models. The running speed of support vector machine is also faster than hidden markov models.Secondly, this paper analyses the influences of error penalty parameter and kernel parameter on the generalization performance of support vector machine in condition of a fixed kernel function, and in the experiments, several groups of error penalty parameter and kernel parameter values were taken to do speech recognition. The experimental results show that the different values of error penalty parameter and kernel parameter affect the generalization performance of support vector machine and accordingly affect the recognition effect of the speech recognition system.In order to meet the need of real-time and portable characteristics of the speech recognition system, this paper proposes a method for speaker-independent word speech recognition system which combines MFCC with SVM on OMAP5912embedded speech recognition system. This system makes the speech recognition more convenient, quickly and universal than traditional speaker-dependent isolated word speech recognition system.