How on-device ML trigger detection works

Dry Fire Pro

The core promise of Dry Fire Pro is simple: press your trigger, and the app knows — within a fraction of a second, with nothing attached to your firearm. Here’s how that works.

The sound of a trigger

A dry trigger press has a distinct acoustic signature: a short, sharp mechanical click as the striker or hammer falls. It’s quiet compared to speech, but it’s consistent — the same firearm produces nearly the same click every time.

That consistency is what makes the problem a good fit for machine learning. A model can learn what trigger clicks sound like across many firearms, and then spot that pattern in a live microphone feed even with room noise around it.

On-device, in real time

Dry Fire Pro runs its detection model directly on your iPhone using Apple’s on-device ML stack. The microphone feed is analyzed in real time and immediately discarded — nothing is recorded, stored, or uploaded. This matters for three reasons:

  • Latency. There’s no round-trip to a server, so a detected press is timestamped within milliseconds — which is what makes measured splits meaningful.
  • Privacy. Your audio never leaves the device. No account, no cloud, no tracking.
  • Reliability. Detection works in airplane mode, in the basement, anywhere.

Universal vs. personalized models

The built-in Universal Gen2 model is trained across a wide range of common handguns and works out of the box for most shooters.

But triggers vary — striker vs. hammer, polymer vs. steel frames, and some triggers are simply quiet. That’s why Dry Fire Pro also includes My Firearm: a guided flow that records a series of your own trigger presses and trains a personalized model on your device, in about a minute. The result is a detector tuned to the exact sound your firearm makes, in the room where you actually train.

Tuning detection

Two settings help detection match your environment:

  • Sensitivity — raise it for quiet triggers or larger rooms, lower it if ambient noise causes false positives.
  • Sound suppression delay — controls how quickly the app re-arms after its own audio cues.

Most shooters never need to touch either. If you do, the combination of sensitivity plus a personalized model handles nearly every setup.

Try it

The best way to understand detection is to run one drill: beep, draw, click — measured. Dry Fire Pro is free on the App Store.

Train smarter with Dry Fire Pro on your iPhone.

Download on the App Store