Table 2 Model comparison.

From: TDNN achitecture with efficient channel attention and improved residual blocks for accurate speaker recognition

Model

EER

minDCF

ACC(%)

Params (M)

TDNN(C=512)39

0.185

0.812

84.4

3.2

ECAPA-TDNN(C=512)9

0.179

0.799

89.1

6.9

Res2Net40

0.218

0.784

90.1

5.6

EResNet41

0.148

0.685

94.5

6.6

ResNetSE42

0.157

0.732

93.6

7.8

EPCNet-TDNN(C=512)

0.154

0.724

95.7

13.9

Back to article page