Show HN: Live breath detection and biofeedback from a phone microphone
A doctor built a breathing app that uses your phone's microphone for live biofeedback — and it runs entirely on-device with ML.
Why it matters
On-device ML for health monitoring is a growing category. This shows how signal processing + machine learning can deliver real-time biofeedback without uploading raw data, a model other health apps may adopt.
The key facts
13 to knowBuilt by Felix, emergency medicine/intensive care doctor from Switzerland
Uses phone microphone for live breath detection and biofeedback
Combines signal processing, breathing state machine, and ML
All processing done on-device — no raw audio uploaded
Tracks rhythm, depth, regularity with focus on self-awareness vs. gamification
Open source on GitHub (shiihaa-app/shiihaa-breath-detection)
Posted on Hacker News with 20 points, 9 comments
On-device ML processing (no data upload)
Signal processing + breathing state machine for real-time feedback
Focus on self-awareness over gamification/scoring
Early-stage project (Show HN post, 20 points, 9 comments)
Built by emergency medicine doctor from Switzerland
Analyzes rhythm, depth, regularity from phone microphone
Go to the source
Hacker Newsgithub.com
Publisher excerpt: Hi everyone, I am Felix, a famliy doctor from ZH, Switzerland. A couple of month ago I started this little project called shii • haa, a breathing app that uses the phone`s microphone for live biofeedback My prior work in emergency medicine and intensive care was closesly linked to breathing, mostly…
