I built AllEars, an Android app that listens for real-world sounds — like snoring, coughing, barking, alarms, glass breaking, and more — and lets you automate actions on your phone using simple IF–THEN flows.
It runs fully on-device, with no cloud, no mic streaming, and no internet requirement. Your phone detects sound events locally using TensorFlow Lite (YAMNet) and executes your defined actions.
Examples:
IF snoring AND coughing THEN vibrate + log timestamp
IF glass break THEN trigger webhook
IF dog bark OR doorbell THEN send notification
Each flow uses flat AND/OR logic, supports multiple actions, and can include time filters. The app runs as a low-power background service with a persistent notification for transparency.
Available actions:
- Play a sound
- Vibrate
- Send a notification
- Log a timestamp
- Trigger a webhook
Why I built it: I wanted user-programmable automation driven by ambient sound — without sending any audio to the cloud.
Think of it as IFTTT for offline sound events.
Try it on Play Store: https://play.google.com/store/apps/details?id=com.cliqueraft... — no signup needed. Free plan includes 3 flows.
Looking for feedback on:
1. What sounds would you want to detect and respond to?
2. What other actions/triggers would make this genuinely useful (keeping things offline and safe)?
3. Any UI feedback on the flow builder logic?
Thanks! Happy to dive into anything technical