User Authentication Method via Speaker Recognition and Speech Synthesis Detection

Table 2

Accuracy by the number of hidden layers and the number of neurons.

# of hidden layers	The number of neurons										Accuracy
	1^st layer	2^nd layer	3^rd layer	4^th layer	5^th layer	6^th layer	7^th layer	8^th layer	9^th layer	10^th layer
0	—	—	—	—	—	—	—	—	—	—	0.59
4	256	256	256	256	—	—	—	—	—	—	0.66
5	256	256	256	256	128	—	—	—	—	—	0.69
6	256	256	256	256	128	128	—	—	—	—	0.73
7	256	256	256	256	128	128	128	—	—	—	0.75
8	256	256	256	256	128	128	128	128	—	—	0.80
9	256	256	256	256	128	128	128	128	128	—	0.79
10	256	256	256	256	128	128	128	128	128	128	0.79