Speaker Gender Recognition Based on Deep Neural Networks and ResNet50

Figure 11

(a) Validation accuracy. (b) Error rate of Resnet34 and ResNet50.

(a)

(b)