Point your phone at a bird, frog, or cricket near you and tap record. The same model running on this land runs right here in your browser and tells you what it is. (This records the world around you; it doesn't play sounds back.)
This is the on-device model running live. It's a broad-region version that covers the species common across your part of the country, not the finely-tuned ear a real Still Habitat install trains on your exact land. It only listens for a fixed set of wildlife, so everyday sounds like voices, a TV, or traffic may come back as a low-confidence guess.
An uncommon visitor one of the ears on this land picked up not long ago.
Every species the ears here have named, most-heard first.
Your land, your model. No servers to run, and no audio leaving the property.
An ear records around the clock and names species right on the device, using a model trained on your own land. No audio ever leaves the property.
All that's published to this page is the result: the species and how sure it is. No recordings, no locations. It updates the moment a bird sings.
Place more ears anywhere on the land. Each one listens and names on its own, so coverage grows and the model keeps learning your place.
Which species rule the daylight hours, and which come alive after dark.
Detection intensity by hour across the most-heard species. The darkest cells are peak activity.
The land wakes up in a particular order.
Every species heard here is logged by an ear: a small, low-power listener that names wildlife on the device, around the clock, with a model trained on your own land. No cameras and no recordings leave the property, only the sound. The busiest stretch is the dawn chorus, when the quiet of the night gives way to a wall of song.
Notes left by guests who stayed and listened.