Feature Extraction and Classification of Emotions in Wave Files Using Crossbreed Algorithm

The importance of automatically recognizing emotions in human speech has grown with theincreasing role of spoken language interfaces in human-computer interaction applications. In this paper,emotion classification method based on hybrid of SVM and HMM algorithm is presented. Four primaryhuman emotions, including anger, aggressive, happiness and sadness are investigated. For speech emotionrecognition, we extracted 15 features to form the feature vector. Extracted features were sent into theimproved crossbreed algorithm (hybrid of HMM & SVM) for classification and recognition. Results showthat the selected features are robust and effective for the emotion recognition and give better accuracycompared to individual SVM & HMM classifiers.