EURASIP Journal on Audio, Speech, and Music Processing

Table 12 Recognition rates of DAE with varying numbers of layers for training data (%)

From: Deep neural network-based bottleneck feature and denoising autoencoder-based dereverberation for distant-talking speaker identification

Method	Number of layers	CENSREC-4 database			Ave.
		0.40	0.60	0.75
DAE	1	95.35	90.65	95.20	93.73
DAE	3	95.40	91.50	94.70	93.87
DAE	5	94.30	90.85	93.80	92.98
DAE	7	89.65	84.40	87.60	87.22
DAE	9	87.85	82.75	87.50	86.03

Back to article page