Automatic Audio Based Emotion Recognition System: Scope and Challenges

Abstract

Emotion is an integral part of human behavior and inherited property in all mode of communication. We, human is well trained thought your experience reading recognition of various emotions which make us more sensible and understandable. But in case of machine, however, it can easily understand content based information such as information in text, audio or video but still far behind to access the depth behind the content. It is need of the era that machine should also be trained to understand emotions correctly for better understanding and to avoid any miscommunication. Present study comes into domain of emotion recognition from audio conversation. Moreover, Audio emotion analysis has many applications in various sectors like healthcare, banking, defense and IT. On the other part, text emotions are easy to decode as there is no role of factors like tone and pitch, but in case of audio emotion analysis both the factors need attention for better accuracy. Also there are several factors like noise, disturbance, and various pauses in communication which results in degrading the accuracy. It is a challenging task to make machine to understand the emotion of the respondents.