Skip to main content
B

BioSonic

BioSonic provides a cloud‑based AI platform that automatically classifies bat species and distinguishes feeding versus social calls from uploaded WAV recordings, delivering species‑level results with 98.9% F1 accuracy. The service includes confidence scoring, an interactive spectrogram viewer, and mapping visualizations to support ecological consulting, government agencies, and research groups in habitat assessment and biodiversity monitoring.

StockholmFounded 202431K+ followers
Updated 3 months ago

Funding

Funding not disclosed

Funding rounds are not available yet.

Founders

Founder details are not available yet.

Product

Problem

Ecological consultants and researchers must manually review thousands of ultrasonic bat recordings to identify species and behavior, a time‑consuming process prone to errors and limited by the need for specialist expertise.

Solution

BioSonic offers a cloud‑based AI platform that automatically classifies bat species and separates feeding versus social calls directly from uploaded WAV files. The system converts recordings into spectrogram images and applies deep neural networks trained on millions of calls, delivering species‑level results with 98.9% F1 accuracy. Users receive confidence‑filtered classifications, behavior insights, and interactive spectrograms for manual verification within a web browser. Integrated mapping tools visualize feeding hotspots and roost locations, supporting habitat protection and impact‑assessment workflows. The platform supports recordings from all major bat detector manufacturers and extracts metadata such as date, time, and GPS coordinates when available.

Target Audience

Primary customers are ecological consulting firms, government environmental agencies, and academic or conservation research groups that conduct bat acoustic surveys for impact assessments, infrastructure planning, and biodiversity monitoring.

Features

  • Automatic species identification for up to 7 species per recording with 98.9% F1 benchmark performance
  • Behavior classification that separates feeding buzzes from social calls
  • Confidence scoring and customizable thresholds to flag uncertain detections
  • Interactive spectrogram viewer for review, override, and audit‑trail tracking
  • Mapping visualizations (heatmaps, pie charts) of feeding activity and roost locations
  • Compatibility with WAV files from all major bat detectors (Song Meter, Pettersson, Batlogger, AudioMoth, Anabat, etc.)
  • Weekly model updates and continuous improvement via retraining on new verified data
  • EU‑hosted storage with AES‑256 encryption and GDPR compliance
This profile is AI-generated and may contain inaccuracies.