Acoustic individual identification in a species of field cricket using deep learning

Individual animal identification is essential for wildlife conservation and management, aiding in estimating abundance and related parameters. The feasibility of identifying individual field crickets (Plebeiogryllus guttiventris) from their calls using deep learning is assessed. In a closed population, the best models recognized individuals with up to 99.9% accuracy when trained and tested on calls from the same night, and 65.1% accuracy when tested on calls from nights not used in training. When matching call pairs without knowing all individuals in the population, models identified pairs of calls from the same night with 97.4% accuracy, falling to 87.8% if calls were from different nights. Accuracy remained high when tested on individuals not observed during training (within night, 95.0%; across nights, 82.7%). Pooling training data from multiple nights improved test accuracy for all models. Deep learning outperformed random forests, especially on harder tasks, although both were able to discriminate individuals. Higher temperatures were associated with shorter chirps and higher frequencies, but adjusting spectrograms for these traits did not improve performance. These results provide the demonstration of acoustic individual identification (AIID), using wild recordings of an insect species, and highlight the potential of deep learning-based AIID for noninvasive animal population monitoring.