Robust Acoustic Emotion Recognition Based on Cascaded Normalization and Extreme Learning Machines

13th International Symposium on Neural Networks (ISNN) -- JUL 06-08, 2016 -- Saint Petersburg, RUSSIAOne of the challenges in speech emotion recognition is robust and speaker-independent emotion recognition. In this paper, we take a cascaded normalization approach, combining linear speaker level, non-linear value level and feature vector level normalization to minimize speaker-related effects and to maximize class separability with linear kernel classifiers. We use extreme learning machine classifiers on a four class (i.e. joy, anger, sadness, neutral) problem. We show the efficacy of our proposed method on the recently collected Turkish Emotional Speech Database.City Univ Hong Kong, Russian Acad Sci, St Petersburg Inst Informat & Automat, IEEE Hong Kong Sect, CIS Chapter, Int Neural Network Soc, Asia Pacific Neural Network Soc, Russian Neural Networks So