F0 and pause features analysis for Anger and Fear detection in real-life spoken dialogs

This paper describes recent work focusing on F0 and pause features detection for two negative emotions, Anger and Fear, occuring in real-life human-human spoken dialogs. Most of the current studies do not differentiate whithin the class of negative emotions, when an automatic system should consider appropriate strategies according to different negative emotions. In this paper we consider two types of prosodic cues aiming to differentiate between two negative emotions Anger and Fear. The work is carried out in the context of the AMITIES project in which spoken dialog systems for call center services are being developed. F0 features are two range parameters, one at the sentence level and the other at the sub-segment level. Pause features are meaningful silent pauses and filler pause “euh”. We correlate all the features with emotion labels and with two variables, gender and speaker (agent vs client). The study shows that pause features are a global more reliable cue to distinguish between Anger and Fear than F0 parameters. However, differences in both F0 and pause patterns needs to be made according to speaker and dialogic context.