Emotion Detection Throughout the Speech

Abstract

Nowadays, technological advances pose new frontiers to society. Artificial Intelligence (AI) has become one of the main research areas of interest due to its enormous possibilities. AI applications are spreading all over various fields, human-computer interaction is no exception. Since the turn of the millennium, the human-machine communication paradigm is shifting to a more efficient way where peripherals such as remote controls, mice or keyboards do not play the center role anymore. Machines are now expected to exhibit a human like behavior, in the sense of detecting, feeling or perceiving human actions, and reacting suitably. Non-intrusive sensing of stress, attention, burnout and emotions for instance, have become possible during human-machine interaction, using AI. In this context, areas such of sensors, images and audio processing and recognition, to name just a few, have been significantly developed. Emotion is an essential part of what means to be human, but it is still disregarded by most technical fields as something not to be considered in scientific or engineering projects. However, the understanding of emotion as an aspect of decision-making processes and of modelling of human behavior is essential to create a better connection between humans and their tools and machines. As voice remains the principal mean of communication of men and is also becoming a usual way of human-machine interaction, detecting emotions throughout speech becomes a powerful toll. In this work, an overview of such issues is done, and a framework to detect emotions throughout speech analysis is presented.