Data-Driven Decision-Support System for Speaker Identification Using E-Vector System

Table 4

Three-utterance experiment micro-F1 score comparison.

Model	Feature	T1	T2	T3	T4	L1	L2	L3	L4
GMM	MFCC	0.990	0.971	0.965	0.973	1.000	1.000	1.000	1.000
	HDCC	0.897	0.793	0.778	0.765	1.000	1.000	0.978	0.992
	E-vector	0.977	0.978	0.975	0.976	1.000	1.000	1.000	1.000
HMM	MFCC	0.940	0.903	0.902	0.895	1.000	0.933	0.925	0.900
	HDCC	0.877	0.827	0.815	0.807	1.000	0.950	0.956	0.958
	E-vector	0.943	0.920	0.905	0.899	1.000	0.950	0.967	0.975