A Review on Emotion Recognition using facial expression and speech

NCIDCAS | Special Issue | December-2018 | Published Online: 22 December 2018    PDF ( 271 KB )
Shinde A.R. 1; Raut S.D. 2; Khanale P.B. 3

1School of computational sciences, Solapur University, Solapur (India)

2School of computational sciences, Solapur University, Solapur (India)

3Dnynopasak College, Parbhani (India)


This paper presents the literature related to the databases, features, pattern classifiers for emotion recognition from facial expression and speech. The interaction between human and computer will be more natural if computer are able to understand emotions of human being. Discrete Cosine Transform (DCT), Discrete Wavelet Transformation (DWT, Fast Fourier Transform (FFT), Singular Value Decomposition (SVD), SUSAN edge detection operator, facial geometry, edge projection analysis, are used to extract the important features for recognizing emotion from facial expression. Hidden Markov Model (HMM) and Artificial Neural Network (ANN) has been proposed to classify emotions. Most of the researcher has be used facial and speech approach for finding human emotions. As per the literature review facial expression gives accurate recognition rate as compare to speech features.

FFT, Discrete Cosine Transform, Discrete Wavelet Transformation
Article View: 16