Table 7.

Average classification performance for the training and testing of the tuned Vgg16 network with data augmentation over 100 training epochs.

Feature	Training tunned Vgg16 with data augmentation		Testing tunned Vgg16 with data augmentation
	Average accuracy	Standard deviation	Average accuracy	Standard deviation
SPEC	0.75	± 0.06	0.71	± 0.04
Chroma	0.75	± 0.10	0.60	± 0.02
MFCC	0.64	± 0.04	0.59	± 0.03
MelSpectrum	0.84	± 0.10	0.67	± 0.03
PowerSPEC	0.85	± 0.10	0.68	± 0.02
RAW	0.77	± 0.09	0.58	± 0.02
Tonal	0.72	± 0.10	0.52	± 0.02
All features	0.76	± 0.10	0.61	± 0.02