Random Deep Belief Networks for Recognizing Emotions from Speech Signals

Table 5

Accuracies (WA) of the compared methods on three databases, illustrating the superiority of RDBN to the other methods.

	L-SVM	KNN	SLDBN	DLDBN	TLDBN	RDBN	BASE
EMODB	81.19	70.74	72.84	53.85	24.59	82.32	79.61
CASIA	42.08	34.33	38.50	29.50	18.25	48.50	36.17
SAVEE	46.25	43.13	30.42	20.62	25.00	53.60	44.58