Speech Emotion Recognition using Deep Neural Networks

The key issues pivotal for successful Speech Emotion Recognition (SER) system are driven by a selection of proper emotional feature extraction techniques. In this book, Mel-frequency Cepstral Coefficient (MFCC) and Teager Energy Operator (TEO) along with a fusion of MFCC and TEO is examined over multilingual database consisting of English, German and Hindi languages. Deep Neural Networks (DNN) has been used for the classification of the different emotions considered, including happy, sad, angry, and neutral. Sample of Matlab code implementation is provided in this book. The proposed system could be implemented especially in the customer service application, in which TEO based features and DNN could be used to better handle customer during a conversation.