My son screams while gaming at midnight. I'm a developer, so I did what developers do - I over-engineered a solution
My son games with headphones late at night. Most of the time he's quiet. Then something happens in the game and he yells at full volume, waking the whole house.
Talking to him about it works for about 20 minutes. Taking the headphones away works but causes a different problem. So I built an app.
S.T.F.U (Sound Trigger Focus Utility) is a Windows tray app that listens to the microphone and interrupts him when he yells. On first run it calibrates by asking him to be quiet, then talk, then yell - so it learns the difference between his normal voice and an actual shout.
When it catches a yell:
- First time that session - it minimizes his game, plays a sound effect, and throws up a fullscreen message he has to click through 4 times. The close button moves after each click so he can't just spam-click one spot.
- Every time after - straight to the desktop with a 10-second message. His game is gone.
Settings are behind a PIN so he can't change the threshold or turn it off. There's a report with a chart showing every trigger, so the next morning I can see exactly when things went sideways.
The important part: he knows it's there. It's not hidden, it's not spyware, the first screen when it launches says exactly what it does. It works because it's a consequence he agreed to, not something I snuck onto his PC. The README explicitly says it works best as something agreed to, not as a hidden trap.
No audio is recorded or stored. It computes a loudness number every 20ms and throws the audio away immediately. Nothing leaves the machine.
It won't solve the underlying issue (he's 100% going to learn to yell more quietly instead of less often), but at least the house sleeps.
Free and open source if anyone wants it: https://github.com/omricn/stfu/releases/latest
