Skip to main content

Table 7 Learning parameters of DNN

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

Batch size 128
Learning rate of Bernoulli-Bernoulli RBM 0.02
Learning rate of Gaussian-Bernoulli RBM 0.002
Weight decay 0.0002
Number of iterations in pre-training 50
number of iterations in fine-tuning 100