Speech Based Emotion Recognition Using MFCC and ANN

Speech is the most natural mode of communication. This work emphasizes on recognizing different emotions from speech signal. There are two major sections in this project namely feature extraction from speech signal and give this features as input to classifier to recognize emotions. Emotional states of speaker are considered as namely angry, happy, sad and neutral. The testing section classifies the training set of data with the help of back propagation algorithm. For the feature extraction of speech signal Mel Frequency Cepstrum Coefficients (MFCC) is used which gives a set of feature vectors of speech waveform. The Artificial Neural Networks (ANN) is selected as the classifier. The whole simulation is taken place in MATLAB environment. The proposed technique is providing promising results by giving the accuracy rate of 80-85 percent. The human capability to recognize the emotion from speech was also studied and compared with classifiers.