Title :
Combining categorical and primitives-based emotion recognition
Author :
Grimm, M. ; Mower, E. ; Kroschel, K. ; Narayanan, S.
Author_Institution :
Inst. fur Nachrichtentechnik (INT), Univ. Karlsruhe (TH), Karlsruhe, Germany
Abstract :
This paper brings together two current trends in emotion recognition: feature-based categorical classification and primitives-based dynamic emotion estimation. In this study, listeners rated a database of acted emotions using the three-dimensional emotion primitive space of valence, activation, and dominance.
Keywords :
emotion recognition; fuzzy logic; acoustic features; categorical-based emotion recognition; direct classification; emotion categories; emotion clusters; emotion primitives classification; emotion primitives estimation; feature-based categorical classification; fuzzy logic classifier; kNN classifier; primitives-based dynamic emotion estimation; primitives-based emotion recognition; speaker-dependent variations; three-dimensional emotion primitive activation space; three-dimensional emotion primitive dominance space; three-dimensional emotion primitive valence space; three-dimensional emotion space; Acoustics; Correlation; Databases; Emotion recognition; Estimation; Speech; Three-dimensional displays;
Conference_Titel :
Signal Processing Conference, 2006 14th European
Conference_Location :
Florence
