EURASIP Journal on Audio, Speech, and Music Processing

Table 7 Accuracy performance of emotion recognition [%] for natural speech

From: Efficiency of chosen speech descriptors in relation to emotion recognition

Clasifier	–	SFS	FMS	S-PCA	SFS + S-PCA	FMS + S-PCA
k-NN	78.9	78.9	80.25	51.30	68.72	77.92
SVM	22.49	22.49	70.91	72.83	78.87	83.95

– without selection, FMS and SFS two different selection methods, S-PCA extraction, SFS + S-PCA and FMS + S-PCA selection and extraction

Back to article page