Deep Learning Analysis Models for Speech and Emotional Recognition

Today speech emotion recognition is one of the major technology usages for every areas in the world. There are many important aspects with speech emotion recognition which would be more convenient and intelligent. Furthermore it is necessary to optimize existing methods to be executable for speech emotion recognition. In this paper, build models based on CNN, SVM and LSTM for speech emotion recognition. Compared with the three methods and analyzed the shallow learning and deep learning models to find a method which has a better performance.