Figure 5From: Deep neural network-based bottleneck feature and denoising autoencoder-based dereverberation for distant-talking speaker identificationGraphic illustration of microphone array.(a) CENSREC-4 (b) RWCP.Back to article page