This publication describes the use of a sound detector to effectively capture and process time series bioacoustic data. According to acoustic surveys done, the data collected informs the development of algorithms used to classify different bird species. This method can be further replicated and combined with a simple classifier to detect cat sounds in domestic recordings.
Title: A changepoint prefilter for sound event detection in long-term Bioacoustic Recordings
Authors: Juodakis Julius, Marsland Stephen & Priyadarshani Nirosha
Journal: The Journal of the Acoustical Society of America
Citation: Juodakis, J., Marsland, S., and Priyadarshani, N., “A changepoint prefilter for sound event detection in long-term bioacoustic recordings”, The Acoustical Society of America Journal, vol. 150, no. 4, pp. 2469–2478, 2021. doi:10.1121/10.0006534.
Open access: No - Credentials required for individual or institutional access
Long-term soundscape recordings are useful for a variety of applications, most notably in bioacoustics. However, the processing of such data is currently limited by the ability to efficiently and reliably detect the target sounds, which are often sparse and overshadowed by environmental noise. This paper proposes a sound detector based on changepoint theory applied to a wavelet representation of the sound. In contrast to existing methods, in this framework, theoretical analysis of the detector's performance and optimality for downstream applications can be made. The relevant statistical and algorithmic developments to support these claims are presented. The method is then tested on a real task of detecting two bird species in acoustic surveys. Compared to commonly used alternatives, the proposed method consistently produced a lower false alarm rate and improved the survey efficiency as measured by the precision of the inferred population size. Finally, it is demonstrated how the method can be combined with a simple classifier to detect cat sounds in domestic recordings, which is an example from the Detection and Classification of Acoustic Scenes and Events (DCASE) 2018 workshop. The resulting performance is comparable to the state-of-the-art deep learning models and requires much less training data.
Key words: Bioacoustics, Data, sound detector, Deep learning models, statistical developments, algorithms
Interested in exploring more on Bioacoustic monitoring? Join our Community's Acoustic Monitoring group for engaging discussions.