Dataset for "Benchmarking for the automated detection of southern yellow-cheeked crested gibbon calls from passive acoustic monitoring data"

"Benchmarking automated detection and classification approaches for long-term acoustic monitoring of endangered species: a case study on gibbons from Cambodia" Recent advances in deep learning and transfer learning have revolutionized our ability for the automated detection of acoustic signals from long-term soundscape recordings. Here, we provide a benchmark for the automated detection of southern yellow-cheeked crested gibbon (Nomascus gabriellae) calls recorded in Jahoo, Cambodia. For the benchmarking, we compared the performance of support vector machines (SVMs), a quasi-DenseNet architecture (Koogu), transfer learning with ResNet50 models trained on the ‘ImageNet’ dataset (ResNet), and transfer learning with embeddings from a global birdsong model (BirdNET). We also investigated the impact of varying the number of training samples on the performance of these models. Transfer learning models based on BirdNET embeddings had superior performance with a smaller number of training samples, whereas Koogu and ResNet models only had acceptable performance with a larger number of training samples (>200 gibbon samples). We deployed the BirdNET-based model over > 130,000 hours of continuous soundscape data, which, after manual review, resulted in >12,000 verified true positive detections. We found that gibbon calling events occurred mostly in the early morning hours between 05:00 to 0:600 local time. We had fewer gibbon detections during the monsoon period and found substantial variation in spatial patterns of calling events across months and years. We show that automated detection can be used to investigate long-term spatial and temporal patterns of gibbon calling events. Reliable automated detection approaches are a critical first step for using passive acoustic monitoring to assess endangered gibbon populations at ecologically relevant temporal- and spatial-scales. Detailed instructions regarding use are provided on GitHub. Link to GitHub: https://github.com/DenaJGibbon/benchmark-gibbon-calls. Please cite both if you use these data: Clink, D., Cross-Jaya, H., Kim, J., Ahmad, A. H., Hong, M., Sala, R., Birot, H., Agger, C., Vu, T. T., Thi, H. N., Chi, T. N., & Klinck, H. (2024). Dataset for "Benchmarking for the automated detection of southern yellow-cheeked crested gibbon calls from passive acoustic monitoring data" [Data set]. Zenodo. https://doi.org/10.5281/zenodo.12706803 Clink DJ, Cross-Jaya H, Kim J, Ahmad AH, Hong M, Sala R, Birot H, Agger C, Vu TT, Thi HN, Chi TN. Benchmarking for the automated detection and classification of southern yellow-cheeked crested gibbon calls from passive acoustic monitoring data. bioRxiv. 2024:2024-08.