Use of Neural Network Classifier for Detecting Human Emotion via Speech Spectrogram

This research presents a novel algorithm for detecting human emotion via speech recognition by using speech spectrogram. The proposed algorithm aims to detect the emotional by using information inside the spectrogram. Neural network was used for being the classifier. A new approach to feature extraction based on analysis of two dimensions time-frequency representation of a speech signal have been presented. The algorithm was tested with EMO-Database. The experimental results show that the proposed framework can efficiently find the correct speech emotion compared to using the traditional methods.