Leveraging tropical reef, bird and unrelated sounds for superior transfer learning in marine bioacoustics

arXiv:2404.16436v1 Announce Type: cross
Abstract: Machine learning has the potential to revolutionize passive acoustic monitoring (PAM) for ecological assessments. However, high annotation and compute costs limit the field's efficacy. Generalizable pretrained networks can overcome these costs, but high-quality pretraining requires vast annotated libraries, limiting its current applicability primarily to bird taxa. Here, we identify the optimum pretraining strategy for a data-deficient domain using coral reef bioacoustics. We assemble ReefSet, a large annotated library of reef sounds, though modest compared to …

abstract annotation arxiv bird compute costs cs.ai cs.lg cs.sd current eess.as however libraries machine machine learning marine monitoring networks pretraining quality transfer transfer learning type vast