Improving Speech Identification with MFCC and SVM

In any language, Spoken alphabet identification as one of the subsets ofspeech identificationand pattern identification has many applications. However, it is not easy to recognize the alphabet. Similar sound that is difficult to detect. One of the problems set is called the E-set that include words letters B, C, D, E, G, P, T, V and Z. This paper describes an approach of speech identification by using the Mel-Scale Frequency Cepstral Coefficients (MFCC) extracted from speech signal of spoken words. The Support Vector Machine (SVM) is used as classifier. In this paper, a method is said to have achieved %08 accuracy on data-set TI ALPHA.