Fig. 6From: Deep semantic learning for acoustic scene classificationThe confusion matrix of average classification results (all-accuracy) under various devices in Mini-SegNet network. a With event extraction block. b Without event extraction blockBack to article page