Papers for

environmental sensor manufacturers

Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.

Bioacoustic classification works across multiple sampling rates accurately

Rate-Agnostic Bioacoustics: Heterogeneous Multi-Taxa Classification with Continuous Filterbanks and Fourier Neural Operators

Abstract: Conventional bioacoustic classification models rely on fixed-rate spectral representations, requiring recordings acquired at heterogeneous sampling rates to be resampled before analysis. We propose a Sampling-Frequency-Independent (SFI) frontend that processes each recording directly at its native sampling rate, coupled with a Fourier Neural Operator (FNO) backbone featuring progressive temporal-scale fusion. This framework avoids fixed-rate resampling and high-frequency information loss while producing fixed-size representations across sampling rates. Mild training-time sampling-rate (\textit{sr}) augmentation further improves robustness to unseen rate variations. Evaluated on a multi-taxa corpus comprising 84 classes and 60 sampling rates, the proposed SFI-FNO configuration outperforms fixed-rate and corpus-maximum-rate baselines, achieving .906 accuracy, .921 balanced accuracy, and a Macro-F1 score of .899.

Tue 29 SeptSound
The gist
Audio recordings of animal sounds often come in different quality levels, which makes it hard for computers to understand them if they rely on a fixed format. The authors developed a system that can learn to identify animal species from their sounds no matter how the recordings were made. Their method processes each recording in its original quality and combines information over time in a smart way. This approach was tested on a large and varied collection of animal sounds and performed better than standard methods that require changing the audio format first.
Open → 2609.37540v1

Hardware friendly bat detector cuts noise and saves battery life

Bad: Taming the Bioacoustic Data Deluge with a Bat Acticity Detector

Abstract: Passive Acoustic Monitoring of bats generates massive ultrasonic datasets (>27 GB/night per node), straining edge storage and battery life. Legacy triggers fail against acoustic confusers, while deep models exceed microcontroller limits. We present a hardware-aware Bat Activity Detector (BAD) specifically designed to discriminate bat calls from hard biological and environmental confusers across variable sampling rates (192-384 kHz). Tailored for the Silicon Labs EFM32PG26 (MVP) in 8-bit integer precision, our model achieves 100 percent hardware offload across all 14 layers (17.2 KB Flash, 73.1 KB RAM). End-to-end preprocessing (74.00 ms for 76 frames) and inference (30.00 ms) of 100 ms clips at 192 kHz require 104.00 ms per clip. On spatially out-of-domain recordings under a realistic low-prevalence regime (r_pos = 0.05), BAD achieves an AUC-ROC of 0.9748 and suppresses 99.4% of non-target noise frames while retaining 65.3% of bat calls - delivering a >33x precision gain over classical Goertzel baselines.

Tue 29 SeptSound
The gist
Passive monitors that record bat calls create huge amounts of data that are hard to store and analyze on small devices. The authors made a new bat call detector designed to run efficiently on a tiny microcontroller, which uses very little memory and processing power. Their detector works well even when tricky noises are present, recognizing many real bat calls while filtering out most other sounds. This improves detection accuracy over older methods without draining device batteries quickly.
Open → 2609.37518v1