Topics for thesis

Emotion recognition is an important part of Human-Computer Interaction (HCI). It has various applications in industrial and commercial sectors. In the context of emotion classification, HCI has been divided into three major modalities: vocal-, facial-, and gesture-based recognition. This title focuses solely on audio based input. Humans use intrinsic skills to infer characteristics denoting the emotional patterns from the voice. However, automation of such a process might prove challenging. In order to identify the emotional state of a given utterance one has to derive relevant features, which carry the emotional information. Finding the most discriminative criterion describing emotional states is sought extensively.