Recognition System for Emotions from Human Speech

Recognition of emotion from human speech is a technology where the emotion or the sentiment is detected from the person's speech. The voice in the speech not only gives us a useful message but also tells us the emotional state of the person speaking. Predicting these emotions is a tough task because different people have different tone and pitch. In this project we have used Mel frequency Cepstral coefficient (MFCC) for feature extraction and Multi-layer perceptron(MLP) classifier for classification of sentiments. The project is mainly to enhance the communication between human and machine for effectiveness and comprehension.