Instantaneous Emotion Detection System using Vocalizations

Abstract

The importance of automatically recognizing emotions from human speech has grown immensely in human computer interaction application. This paper describes an experimental study on the detection of emotion from speech. The study utilizes a corpus containing emotional speech with 25 short utterances expressing three emotions: anger, happiness and neutral (unemotional) state, which were captured manually from recording. Emotions are so complex that most speech sentences cannot be precisely assigned into a particular emotion category; however, most emotional states nevertheless can be described as a mixture of multiple emotions. Based on this concept some of the acoustic features are extracted and the voice samples are trained using Locality Preserving Projection method and Kullback-Leibler Distance to recognize utterances within these three categories. Before detecting the emotion initially Kalman Denoise filtering technique is applied in order to remove noise. This paper is a model of Emotion Detection System in order to understand how emotional viewing could be coded and that has led us to the devising of emotional film structures. With this work both the negative and non-negative emotions can be easily detected.