Room adaptive conditioning method for sound event classification in reverberant environments
This addresses robustness issues for sound event classification systems in real-world indoor settings, though it is incremental as it builds on existing conditioning approaches.
The paper tackled performance degradation in sound event classification caused by reverberation in indoor environments by proposing a conditioning method that uses room impulse response (RIR) information to make networks less sensitive to environmental factors, resulting in reduced performance degradation.
Ensuring performance robustness for a variety of situations that can occur in real-world environments is one of the challenging tasks in sound event classification. One of the unpredictable and detrimental factors in performance, especially in indoor environments, is reverberation. To alleviate this problem, we propose a conditioning method that provides room impulse response (RIR) information to help the network become less sensitive to environmental information and focus on classifying the desired sound. Experimental results show that the proposed method successfully reduced performance degradation caused by the reverberation of the room. In particular, our proposed method works even with similar RIR that can be inferred from the room type rather than the exact one, which has the advantage of potentially being used in real-world applications.