Acoustic AI Detection
Automated detection and classification of bat, bird, and rare-event vocalisations in the field and in archived recordings.
We run classifiers on field hardware when decisions need to happen immediately, and in the cloud when large archives need to be processed quickly. Field detection can work without reliable internet and can trigger alerts within seconds. Cloud processing is used for retrospective surveys, classifier comparison, and rare-event searches across years of recordings.
Technology it uses
- Ultrasonic Microphones High-bandwidth (sub-1 Hz to >190 kHz) microphones for recording bat
- Edge ML Inference Lightweight ML classifiers running on field hardware — classifying
- Bulk ML Inference High-throughput cloud GPU pipelines for processing archived
- Field Platform The physical and electronic base that lets every monitoring system
- Dashboards & Review Tools Client dashboards provide access, replay, and report downloads.
Questions this system helps answer
- Presence and Absence Is the target species present at this site, and when?
- Abundance and Colony Size How many animals use this site, cave, corridor, or roost?
- Movement and Connectivity How do individuals move between roosts, caves, or across landscapes?
- Flight Height and Collision Risk At what heights, times, and conditions do animals fly through wind energy infrastructure?
- Roost Discovery and Use Where are roosts, which are active, and how is roost use changing?
- Disturbance and Displacement How do animals respond to construction, mining, exclusion events, or novel structures?
- Breeding and Seasonality When do animals breed, and how does site use vary through the year?
- Noise, Vibration, and Compliance Are noise and vibration levels within regulatory thresholds at sensitive receivers?
Projects that have used this
Want to use this system?
We tailor each deployment to the project. If this system fits what you need to measure, get in touch.