In any language, Spoken alphabet recognition as one of the subsets of speech recognition and pattern recognition has many applications. The purpose of audio signal processing, they are classified. Speech recognition is one of the issues in computer science and artificial intelligence, which seeks to identify a person based on the person's voice. Alphabet Recognition Speech recognition is below the branches. The different methods of feature extraction and classification, in this paper, a method combining these algorithms are trying to improve the English alphabet recognition. We are also leading to problems such as these problems can be noted that E-set, this collection contains the letters B, C, D, E, G, P, T, V and Z. The problem is similar to the set of waves vocal alphabet E that makes it difficult to recognize in all this is set in this paper by using MFCC feature extraction and SVM classification methods to achieve our desired results. In this paper, a method is said to have achieved 80% accuracy on data-set TI ALPHA.