EmoLogus: Presentation of a linguistically based model for emotion detection and adaption to another context

This paper presents a system which aims at characterizing emotions in speech by only considering linguistic content. It is based on the assumption that emotions can be compound: simple lexical words have an intrinsic emotional value, while verbal and adjectival predicates act as a function on the emotional values of their arguments. The paper describes the compositional computation algorithm of the emotion, as well as the lexical emotional norm used by this algorithm. A quantitative and qualitative analysis of the differences between system outputs and expert annotations is given, which shows satisfactory results, with a good detection of emotional valency in 90.2% of the test utterances. After that, a description of the adaptation of the system is presented.