Skip to content

AwareFlow™ — A privacy-first iOS app that detects unconscious audible habits like sniffing and throat clearing using on-device machine learning. Private nudges help you notice before someone else points it out. No audio ever recorded.

Notifications You must be signed in to change notification settings

getsnaphabit/awareflow-static

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

474 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

title colorFrom colorTo sdk app_file license custom_domain short_description description date_modified
AwareFlow™
blue
indigo
static
index.html
other
awareflow.app
Build awareness of your habits — gently, privately, on your terms.
AwareFlow™ is a privacy-first iOS app that helps people notice unconscious audible habits like sniffing and throat clearing using on-device AI. A complete awareness practice — not just a detector. Gentle nudges, pattern insights, a Reflection Journal, and a Response Toolkit. No audio ever stored or transmitted. Awareness built through curiosity heals. Surveillance built on shame harms.
2026-02-23

AwareFlow™

Build awareness of your habits — gently, privately, on your terms.

AwareFlow helps people become aware of unconscious audible habits — sniffing, throat clearing, and the sounds we don't realize we're making — before they affect the relationships we care about. Using on-device machine learning, AwareFlow delivers a complete awareness practice: real-time detection, private nudges, personalized calibration, pattern insights, a Reflection Journal, and a Response Toolkit of gentle alternatives. All audio is processed locally and never stored or transmitted. Privacy isn't a feature — it's the architecture.


Why AwareFlow Exists

Millions of people have unconscious audible habits they don't notice — but the people around them do. For partners and family members, especially those living with misophonia, these small sounds can become a constant source of tension. The silence around these habits creates shame on both sides: one person feels nagged, the other feels unheard.

AwareFlow was built for both of them. It gives the person making the sound a way to notice on their own — with curiosity instead of criticism, and awareness instead of anxiety. The partner gets relief without having to be the one who points it out.

The app is awareness-first, gentle by design, and grounded in research on Habit Reversal Training, Acceptance and Commitment Therapy, and trauma-informed design. It is not corrective, punitive, or behavior-policing. AwareFlow exists to create space, not pressure.

Read the founder's story: awareflow.app/story


What AwareFlow Does

A complete awareness practice — not just a detector.

Real-time habit detection — Custom-trained sound classifiers identify sniffing and throat clearing using on-device machine learning. No audio is ever recorded, stored, or sent anywhere.

Private nudges — A gentle vibration or quiet notification, visible only to you. AwareFlow calls these Intentions, not goals, because they can't be failed. You choose how, when, and how often you'd like to be nudged.

Calibration Lab — Builds a personalized sound profile for your environment, measuring background noise and tuning detection to your specific space. Recalibrate anytime your surroundings change.

Context and emotion — Each moment is enriched with time of day, ambient noise, weather conditions, and schedule density. Optional mood check-ins connect how you're feeling to when habits show up — surfacing patterns you'd never notice on your own.

Reflection Journal — Browse past reflections grouped by day, search by feelings or notes, and share meaningful moments with someone you trust. Your journal tells a story of growth — not a record of what went wrong.

Response Toolkit — When you notice an urge, a toolkit of gentle alternatives: breathing techniques, sensory exercises, and body-based responses. These are choices from your toolkit, not corrections.

Learning Journey — A structured series of lessons on the science and practice of building awareness — written in plain, warm language. No clinical terminology. No homework.

Gentle Insights — Patterns, not scorecards. "More active than usual" instead of raw counts. Context instead of judgment. Direction instead of numbers. Your streak counts days you showed up — rest maintains it. There is no way to break it.

On-device AI coaching — Powered by Apple Foundation Models (iOS 26+), insights are generated entirely on your device. No cloud, no data sharing.


The Philosophy

Awareness built through curiosity heals. Surveillance built on shame harms.

That belief shapes everything:

  • We say "pattern," never "symptom." We say "moment," never "incident." We say "intention," never "goal."
  • Your data tells a growth story. We show patterns instead of scores. Context instead of judgment.
  • Noticing is the practice. Reduction may follow naturally, but it's a side effect of awareness — not the goal.
  • Rest is part of the practice. Your streak counts days you showed up — and showing up includes taking a break. There is no way to fail.
  • The app is a mirror, not a judge. A companion, not a clinician. A journal, not a ledger.

We drew from Habit Reversal Training, Acceptance and Commitment Therapy, trauma-informed design, and Gentler Streak's Apple Design Award-winning approach to rest-positive fitness. Then we made it our own.


Privacy First

Privacy is foundational — not a feature.

  • Microphone access is session-based, not constant
  • Detection runs fully on-device via CoreML and AVAudioEngine
  • Raw audio is analyzed in memory and immediately discarded
  • No servers, no third-party analytics, no data sharing
  • Zero advertising or tracking identifiers
  • Production builds have zero detection-related console output

We literally cannot hear you. The technology makes it impossible.

Read more: awareflow.app/privacy


Patent-Pending Technology

U.S. Provisional Patent Application No. 63/924,802 Filed November 25, 2025

Systems and Methods for On-Device Acoustic Habit Detection and Context-Aware Behavioral Insight Generation Using Personalized Sound Signatures and Multi-Source Metadata

Four patent pillars: on-device detection, personalized calibration, contextual metadata enrichment, and an on-device insights engine.


Tech Stack

  • Swift 6.2+ with strict concurrency (iOS 26.3+)
  • SwiftUI-only architecture — no UIKit views
  • CoreML + AVAudioEngine for on-device acoustic detection (CPU-only inference)
  • SwiftData for local persistence, CloudKit for optional sync
  • StoreKit 2 for Apple-managed commerce
  • Apple Foundation Models for on-device AI coaching
  • Structured logging, localization, and full accessibility (VoiceOver, Dynamic Type, Reduce Motion)

Explicitly not used: Firebase, RevenueCat, Combine for app logic, third-party UI frameworks, or any external analytics.


Current Status

AwareFlow is in TestFlight with external beta testers. Two habit classifiers are trained and active (sniff and throat clearing), with five more planned. The Gentler System — a curiosity-based awareness philosophy grounded in HRT, ACT, and trauma-informed design — is the foundation for all user-facing messaging, copy, and interaction design.


Learn More


License

This is a proprietary project. All rights reserved by SnapHabit LLC. UI copy, branding, detection models, and trademarks are proprietary to SnapHabit LLC.


Author

Jason Babcock, MBA, ACRP-CP Founder, SnapHabit LLC Certified Clinical Research Professional


Built with empathy. Designed with curiosity. Shipped with care.

About

AwareFlow™ — A privacy-first iOS app that detects unconscious audible habits like sniffing and throat clearing using on-device machine learning. Private nudges help you notice before someone else points it out. No audio ever recorded.

Topics

Resources

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •