A review of automated emotion recognition points to non-contact technologies

Together these developments in automated technology for the detection of emotion and developing understanding of social signals processing raise interesting research questions regarding the use that can be made of these within various scenarios. The specific research proposed here focusses upon training for interpersonal tasks, with the initial use-case of media skills training (that is training to give an effective TV or radio interview). The aim of the research will be to explore the impact of a training intervention based on automated recognition of human emotion and non-verbal behaviour. The project will use a number of technologies designed to recognise emotional expressions, body movement and physiological signals with initial exploratory work focussing on which combinations are most promising for predicting 'good' interview performance. Later experimental work will compare training interventions with and without affect-based intervention for the chosen contexts of use.