Speaker Gender Recognition Based on Deep Neural Networks and ResNet50

Table 3

The classification performance on the test set for different models.

Features	Model	Accuracy (%)
MFCC, Mel spectrogram, Chroma STFT, Tonnetz, special contrast	Designed DNN	95.97
	MLP	95.81
	neighbor classifier	95.10
	Random forest classifier	94.23
	SVC RBF kernel	93.92
	SVC	91.63
	Ada boost classifier	90.13
	Decision tree classifier	88.70
	Quadratic discriminant analysis	77.33
	Gaussian NB	72.27