Skip to content

Dataset · geronimobasso

Drone Audio Detection Samples

geronimobasso/drone-audio-detection-samples

A large open collection of drone audio samples (16 kHz, mono) for training acoustic detection systems.

What's inside

The Drone Audio Detection Samples (DADS) dataset contains between 100k and 1M audio files and is positioned as the largest open UAV sound database. All recordings are standardized to mono, 16 kHz, 16-bit, with durations from 500 milliseconds to several minutes. Most drone sound clips are manually trimmed to guarantee the target signal is present in the recording.

What you can do with it

  • Train sound classifiers for acoustic drone detection systems
  • Train lightweight audio models for UAV detection on edge devices (Edge AI)
  • Test the robustness of monitoring systems against false positives from background noise
Tasks
Audio classification
Languages
English
Size
100K–1M rows
Downloads this month
2.2K
Likes
42

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("geronimobasso/drone-audio-detection-samples")