Table 1.

Literature review on emotion recognition using physiological and speech signals.

Ref. No.	Signals	Features	Classifiers	Emotion Parameters	Stimuli	No. of Subjects	Accuracy in %
[11]	EMG ECG EDA RSP	Statistical, Energy, Sub band Spectrum, Entropy	Linear Discriminant Analysis	Joy, Anger, Sad, Pleasure	Music	3 , MITdatabase	95 (Subject-Dependent) 70 (Subject-Independent)
[14]	EDA HR ST	No specific features stated	KNN, Discriminant Function Analysis, Marquardt backpropagation	Sadness, Anger, Fear, Surprise, Frustration, Amusement	Movies	14	91.7 (Subject-Dependent)
[18]	EMG EDA BVP ECG RSP	Running mean Running standard deviation Slope	NN	Arousal, Valance	IAPS (Visual Affective Picture System)	1	96.58 Arousal 89.93 Valence (Subject-Dependent)
[12]	ECG	Fast Fourier	Tabu Search	Joy, Sadness	Movies	154	86 (Subject-Independent)
[20]	EDA HR	No specific features stated	fuzzy logic	Stress	Hyperventilation Talk preparation	80	99.5 (Subject-Independent)
[19]	BVP EMG ST EDA RSP	Statistical Features	SVM, Fisher LDA	Amusement, Contentment, Disgust, Fear, Sad, Neutral	IAPS	10	90 (Subject-Dependent)
[9]	EMG EDA ECG BVP ST RSP SPEECH	Statistical Features, BRV, Zero-crossing, MFCCs	KNN	Arousal, Valance	Quiz dataset	3	92 (Subject-Dependent) 55 (Sub Independent)
[21]	EDA HR EMG	No specific features stated	HMM	Arousal, Valance	Robot Actions	36	81 (Subject-Dependent) 66 (Subject-Independent)
[13]	EDA ECG ST	Statistical Features average power SCL SCR	CNN	Arousal, Valance	Movies	10	82.35 (Subject-Independent)

EMG: Electromyography; ECG: Electrocardiography; EDA: Electrodermal Activity; RSP: Respiration; ST: Skin Temperature; EEG: Electroencephalogram; BVP: Blood Volume Pulse; HR: Heart Rate; KNN: k-nearest neighbors algorithm; SVM: Support vector machine; HMM: Hidden Markov Model; ANN: Artificial Neural Network; CNN: Cellular Neural Network.