Skip to main content

Table 11 Recognition rates of DAE with different sized contexts for the training data (%)

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

Method Context size CENSREC-4 database Ave.
   0.40 0.60 0.75  
DAE 9 (c1+p8) 94.30 90.85 93.80 92.98
DAE 9 (c1+p4+l4) 93.40 89.60 93.15 92.05
DAE 9 (c1+l8) 93.50 89.65 92.75 91.07