Jump Function Kolmogorov for overlapping audio event classification

This paper presents a novel method for audio event classi cation in overlapping conditions. The method is based on Jump Function Kolmogorov (JFK), a stochastic representation, which is (a) additive, thus the sum of signal and noise yields the sum of their JFKs; (b) sparse, therefore audio events are separable in this domain. The proposed method is an extension of our previous works for classication under noise-mismatch conditions. Similar to that approach, the robustness of the JFK feature is obtained by limiting them within con dence intervals, which can be learned in advance. However, in order to classify overlapped events, we design the classi cation system as a set of event detectors and develop a novel approach which maps JFKs to a speci c feature for each detector. The experiment shows that the proposed method achieves promising results in very challenging overlapping conditions.

• Type: Journal
• Year: 2011
• Where: ICASSP
• Authors: Tran Huy Dat, Haizhou Li
• Keywords: Classi Cation, Event Classi Cation, ICASSP 2011, Overlapping Conditions, Signal Processing