Drop an audio or text file in your own storage bucket. Earshot plays it on your Android phone through whatever you're listening on: car, earbuds or speaker. It waits until your calls end and never plays a stale message.
A message that arrives mid-call is held, then played from the start once you hang up. Call time doesn't count against its age limit.
Every message has an age limit, 30 minutes by default. A message older than that is marked missed instead of being read out to you an hour later.
Each message records detected, started, finished or missed, with the reason and the output it played on. Send those receipts to your own endpoint and your agent can check them.
Messages play automatically over Bluetooth, the steering-wheel buttons skip, and the full text is on screen when you're parked.
Anything dropped in alarm/ plays through silent mode and Do Not Disturb, even during a call, then puts your volume back where it was.
Drop a .say file and the phone reads it in its own voice, with no text-to-speech server to run.
Point Earshot at your bucket. Any S3-compatible bucket works: Cloudflare R2, MinIO, Backblaze B2 or AWS S3. Give the phone a read-only key.
Drop a file from anything that can upload one: a coding agent's hook, a CI job, n8n or Home Assistant.
echo "Build finished. Three tests failed." > done.say
aws s3 cp done.say s3://my-earshot/phone/
Hear it. The phone picks it up within seconds and plays it, or holds it until your call ends.
Android 8.0 or newer. There's no iPhone version.
Not yet. The beta is installed directly from this page, and a Play version comes later.
It checks your bucket on a timer while it runs, and that costs some battery. The beta checks every 5 seconds by default, and you can set a slower interval.
Yes, for the beta. A bucket you control keeps your messages private and keeps the price a one-time $4.99.