Using an Automated Speech Emotion Recognition Technique to Explore the Impact of Bullying on Pupils Social Life

Most of the pupils are witness bullying behavior at school. The variety of the roles that bystanders play in school bullying should be seen as a chance of promoting anti-bullying strategies and interventions. The authors suggest the usage of an automated speech emotion recognition technique as part of a multiple methods tool-box of exploring bullying, so as, teachers to identify the nature of the psychological impact experienced by pupils participating in bullying, and to intermediately respond to serious bullying episodes. In this paper we present a speech emotion recognition technique using K Star classifier. Berlin Emotional Database was used for the experiment. This work focuses on speaker and utterance dependent and independent framework. In the speaker dependent framework, K STAR achieves accuracy 77%, while in the speaker independent framework the classification rate reaches 74%. Finally, it is assumed that teachers understanding bystanders´ emotions, and consequently, their coping styles in bullying episodes, they could introduce strategies that give pupils who witness bullying positive roles to counter bullying.