bioacoustics

MIT Technology Review examines how AI and new sensor technologies are helping researchers decode animal communication, highlighting efforts like the Earth Species Project's NatureLM, while noting significant challenges in translating non-human signals and skepticism about over-reliance on acoustic similarity.

Scientists used computer-generated renditions of 1960s rufous-collared sparrow songs to successfully reteach lost birdsong dialects to local birds in Buenos Aires, showing synthetic audio could help conserve dwindling songbird populations whose learned cultural songs have disappeared.

Researchers at Naturalis Biodiversity Centre and Tilburg University investigated whether variation in the interpulse interval (IPI) of bat calls contributes to species classification, comparing EfficientNet-B0 and PaSST with pretrained BatDetect2 and BAT. They found limited support for IPI-based discrimination and little evidence that transformers use temporal context better than CNNs, though models trained on natural IPIs transferred worse to normalized data.

This review article synthesizes active learning research for biodiversity monitoring, addressing label efficiency and highlighting the need for methods that support validation and reliable ecological inference.

CodaBridge is a tool that enables users to listen to sperm whale recordings, shape synthetic codas, compare timing patterns, and use AI to investigate evidence in Context Lab for traceable data exploration.

The paper proposes a framework for processing and classifying invasive bird species vocalizations using Bayesian wavelet shrinkage and supervised learning models, with SVM achieving high accuracy in classification.

ChiroEcho is a deep learning framework that extends automated bat vocalisation classification by combining species and genus predictions with geographic data, increasing operational coverage of European bat species from 73% to 85%.

The U.S. Navy's Cold War-era underwater listening system SOSUS, designed to track submarines, accidentally discovered whale songs and later provided invaluable data for cetacean research.

Seeing Birdsong is a framework that transforms avian vocalizations into geometric forms and 3D visualizations, bridging art and acoustic science. It uses spectral descriptors to create data-rich structures for research, education, and artistic performance.

DeepMind announces new AI research applications for conservation, including a high-resolution deforestation risk model, species distribution mapping using Graph Neural Networks, and updates to the Perch bioacoustics model.

Google DeepMind released an updated version of Perch, an AI model for bioacoustic analysis that helps conservationists monitor endangered species through audio data. The new model improves bird species prediction, adapts better to underwater environments, and expands to include mammals, amphibians, and anthropogenic noise, with over 250,000 downloads of the original version.