This paper proposes an effective scheme for Speech Emotion Recognition (SER),which uses the decision of emotion recognition to combine emotion classes based on speech recognition.Decision-making emotion classes can provide the correct target among base classifiers.In order to demonstrate the high effectiveness of the suggested approach,we recognizing was performed on emotional Berlin Database consisting of seven emotions classes using Probabilistic Neural Network (PNN) classifier,and the set of feature extraction used Mel frequency cepstral coefficients (MFCC).Our studies suggest that the decision-making systems yield behaviours that,to a great extent,resemble what humans perceived and their performance is close to that of humans under the same experimental setup.