Sound Tags

This is the catalog of sound tags the Edge SDK can detect via Sound Event Detection.

Heads-up for Edge SDK projects: unlike the Cloud API, an Edge SDK project only ships the tags you check at project creation. Pick from this catalog when creating the project on the Dashboard—the on-device model is built for exactly that subset. You can update the selection later via the project’s tag list; the SDK will pull the new model on next initialization.

Current deep learning model: v2.40.0(Jun 01, 2026)

  • All
  • Emergency
  • Human action
  • Human status
  • Human speech
  • Home context
  • Outdoor context
  • Animal
  • Music

NOTE: Most sound tags are optimized to be recognized by real sound, not the sound played by laptops, smartphones, or speakers.

If the sound you need isn’t here, train your own via Custom Sound: Sound Events — Edge SDK only.