Emotional speaker recognition in real life conditions using multiple descriptors and i-vector speaker modeling technique Asma Mansour 1 & Farah Chenchah 1 & Zied Lachiri 1 Received: 9 November 2017 /Revised: 25 May 2018 /Accepted: 6 June 2018 # Springer Science+Business Media, LLC, part of Springer Nature 2018 Abstract Emotional speaker recognition under real life conditions becomes an urgent need for several applications. This paper proposes a novel approach using multiple feature extrac- tion methods and i-vector modeling technique in order to improve emotional speaker recog- nition under real conditions. The performance of the proposed approach is evaluated on real condition speech signal (IEMOCAP corpus) under clean and noisy environments using various SNR levels. We examined divers known spectral features in speaker recognition (MFCC, LPCC and RASTA-PLP) and performed combined features called MFCC-SDC coefficients. The feature vectors are then classified using the multiclass Support Vector Machines (SVM). Experimental results illustrate good robustness of the proposed system against talking conditions (emotions) and against real life environment (noise). Besides, results reveal that MFCC-SDC features outperforms the conventional MFCCs. Keywords Speaker recognition . Emotion . I-vector . MFCC-SDC . SVM . Noise 1 Introduction Emotion is a conscious mental reaction accompanied by physiological and behavior changes in human body. Thus, it is very interesting to understand emotions in human communication [21]. Speech is the natural way to convey human feeling. Speech signal contains much Multimed Tools Appl https://doi.org/10.1007/s11042-018-6256-2 * Asma Mansour asmamansour86@gmail.com Farah Chenchah farahchenchah@yahoo.fr Zied Lachiri lachiri.z@gmail.com 1 National school of engineering of Tunis, LR SITI laboratory, University of Tunis El Manar, BP. 3, le belvedere, 1002 Tunis, Tunisia