RECURRENCE QUANTIFICATION ANALYSIS FEATURES FOR AUDITORY SCENE CLASSIFICATION

This extended abstract describes our submission for the scene classification task of the IEEE AASP Challenge for Detection and Classification of Acoustic Scenes and Events. We explore the use of Recurrence Quantification Analysis (RQA) features for this task. These features are computed over a thresholded similarity matrix computed from windows of MFCC features. Added to traditional MFCC statistics, they improve accuracy when using a standard SVM classifier.