Predictive Tool for Peer Support and Mental Health Improvement with Speech Emotion State Classification for University Students

Abstract

Mental health issues such as stress, depression, anxiety and trauma are some of the most common disorders which is mainly affecting many university student’s academics and their future careers. These mental health conditions are quite prominent in the youth generation due to the workload that they carry out during their studies. There are many mental health innovations and self-help applications have been developed for various purposes, but these applications mainly focus on online therapy and communication via video conferences. It has also been reported that these online therapy applications have not been catering to improving the mental status of university students. Also, the recent studies have proved that the peer support is an effective approach to the mild conditions of stress, depression and anxiety. Therefore, the aim is to provide a tool to analyze the mental health state using speech emotion classification for the students and support them by monitoring the emotion variations in order to identify the progress. The proposed solution consists of a live peer support sessions application and a predictive tool for mental health improvement with speech emotion state classification using convolutional neutral networks for the university students.