Automatic Gender Detection Based on Characteristics of Vocal Folds for Mobile Healthcare System

Table 6

Results of gender detection for existing and proposed systems with clean speech.

Performance measures	Existing system			Proposed system
	4 Gaussians	8 Gaussians	16 Gaussians
TP (%)	99.13	99.71	99.71	100.00
TN (%)	91.66	91.66	95.11	96.55
FP (%)	8.33	8.33	4.89	3.45
FN (%)	0.86	0.29	0.29	0.00
SE	0.99	1.00	1.00	1.00
SP	0.92	0.92	0.95	0.97
ACC (%)	95.4	95.68	97.41	98.27
AUC	0.9510	0.9734	0.9795	0.9845