A semi-automatic tool for emotion corpora annotation

The annotation of large corpora is an important aspect in multimodal spoken language dialogue applications where mostly stochastic models need to be trained and evaluated. Especially in emotion recognition and modelling, corpora annotation represents an essential phase. The time required for labelling such corpora is considerable, and manageablity becomes more arduous and tedious. We propose a semi-automatic tool that will help corpora annotators in managing and annotating large sample emotion corpora.