Fig. 8From: End-to-end speech emotion recognition using a novel context-stacking dilated convolution neural networkVisualisations of the local contexts of the testing sets from the IEMOCAP foldsBack to article page