Title :
Recognition of emotion using non-linear dynamics of speech
Author :
Harimi, Ali ; Shahzadi, Ali ; Ahmadyfard, Alireza
Author_Institution :
Dept. of Electr. & Comput. Eng., Semnan Univ., Semnan, Iran
Abstract :
Recognition of human´s emotion from speech has become one of the most challenging and attractive fields of research in speech processing area. The present study aimed to detect valence of emotions, using Non-Linear Dynamic features (NLDs). NLDs are extracted from the Discrete Cosine Transform (DCT) of descriptor contours computed from Phase Space Reconstruction (PSR) of speech. These features are used to estimate emotion primitives in 3D continuous emotion space on the VAM database using Support Vector Regression (SVR) under the Leave-One-Out (LOO) testing technique. Feature selection is performed using Sequential Forward Selection (SFS). Activation, valence and dominance of emotions are estimated by the best correlation of 85.06%, 53.39% and 84.68%, respectively.
Keywords :
correlation methods; discrete cosine transforms; emotion recognition; feature extraction; feature selection; regression analysis; speech recognition; support vector machines; 3D continuous emotion space; LOO testing technique; Leave-One-Out testing technique; NLD feature extraction; PSR; SFS; SVR; VAM database; correlation method; descriptor contour DCT; discrete cosine transform; emotion activation; emotion dominance; emotion valence detection; feature selection; human emotion recognition; phase space reconstruction; sequential forward selection; speech nonlinear dynamic feature extraction; speech processing; support vector regression; Correlation; Correlation coefficient; Databases; Emotion recognition; Feature extraction; Speech; Speech recognition; Non-Linear Dynamic features; Phase Space Reconstruction; Speech emotion recognition;
Conference_Titel :
Telecommunications (IST), 2014 7th International Symposium on
Conference_Location :
Tehran
Print_ISBN :
978-1-4799-5358-5
DOI :
10.1109/ISTEL.2014.7000745
