One-Class Bioacoustic Detector for Monitoring the Critically Endangered Pied Tamarin (Saguinus bicolor)

The pied tamarin (Saguinus bicolor) is a critically endangered primate with a small geographic range that includes fragmented urban forest mosaics in Amazonia, where habitat subdivision and anthropogenic actions complicate its survival and monitoring. Passive acoustic monitoring (PAM) offers a convenient, noninvasive way to track this species, yet open-set rainforest soundscapes make single-species detection challenging. We present a machine-learning pipeline with a very low false-positive rate, appropriate for downstream inference. The method combines a band-pass filter (5 kHz to 10 kHz), Perch bioacoustic embeddings (deep learning), and a One-Class SVM (OCSVM) applied to sliding windows of continuous audio recordings to detect S. bicolor calls. We train on a reduced dataset of labeled calls and validate against diverse out-of-class audio (birds, anurans, anthropophony, and geophony/insects), then test on long, cross-site recordings. The approach achieves high discrimination on held-out negatives and produces very low false-positive rate in continuous, real-world audio, with a precision of 0.86. Finally, we pair detections with a single-site occupancy model in a cross-site setting to illustrate end-to-end utility for conservation monitoring and to estimate the false-negative detection probability in recordings from pied tamarin populations in a different geographic region. Our strategy provides a tool for PAM of S. bicolor that requires minimal manual labeling effort and can be adapted to other open-set, single-species monitoring scenarios. We grant reproducibility by releasing a Python package (sauim-detector), installable via pip, that processes an audio file and produces detection timestamps as an Audacity label file (.txt), enabling faster manual verification. Open-set bioacoustic detector for the pied tamarin. We introduce an open-set pipeline—band-pass (5 kHz to 10 kHz) → Perch embeddings → one-OCSVM—tailored to S. bicolor, filling a gap with no prior species-specific detector. Bird-trained Perch embeddings transfer to primates. We show that Perch embeddings trained on birds generalize to S. bicolor vocalizations, enabling cross-taxon reuse without retraining. Band-pass filtering reduces background false positives and improves separability. On our evaluation sets, the 5 kHz to 10 kHz filter removed all background FPs and shifted ROC curves up/left, increasing AUC. Robust cross-site detection on long recordings with very few false positives. In an approximate 10 min Mindú Park test, the detector prioritized precision, with an observed false-positive rate of 0.03, and improved AUC from 0.74 (raw) to 0.83 (filtered). End-to-end monitoring via occupancy modeling. We couple detections with a single-site occupancy estimator and derive a closed-form MLE to estimate the cross-site false-detection probability. Software package. We release the open-source Python package, which runs locally to apply our pipeline to an audio file and outputs Audacity-compatible detection labels plus a filtered audio file, streamlining manual verification and reducing effort on large PAM datasets (https://pypi.org/project/sauim-detector).