AURA

Your data forgets why it mattered. AURA captures emotional context and compresses it into tiny Base44 codes that revive the meaning behind your digital and human memories—privately, on-device.

Comment

Inspiration

We save links, notes, screenshots, reminders — but later we forget why we saved them. The emotional context disappears. The same happens in human interactions — we forget important personal details that matter to relationships. AURA was inspired by the idea that memory is not just data — it’s emotion, intention, and meaning.

What it does

AURA captures emotional context at the moment of saving or interacting, and compresses it into 5-character Base44 codes. These tiny codes act as anchors to the human story behind our data — letting us instantly recall what mattered and why.

  • Save a file → Base44 stores emotional + situational context
  • Meet someone → AURA remembers personal details that matter
  • Scan a code → AURA revives the moment

All processing is on-device, private, and stored as Base44 — not raw conversations.

How we built it

We designed a lightweight architecture:

  • Local NLP model for emotional signal extraction
  • On-device inference using a privacy-first approach
  • Custom Base44 encoding system
  • Context-embedding and compression
  • Prototype UI for file tagging & contact memory hints

We focused heavily on user trust and data privacy — no cloud storage, no external logging, no tracking.

Challenges we ran into

  • Designing Base44 encoding to fit complex context into 5 characters
  • Preventing accidental over-capture of data
  • Balancing emotional inference with privacy
  • Avoiding creepy “surveillance-like” behavior
  • Creating a UX that feels magical rather than intrusive

Accomplishments we’re proud of

  • We turned memory into something scannable.
  • We transformed raw emotion into a compact digital artifact.
  • We built a system where your memories stay yours — not on someone else’s server.

What we learned

  • Privacy is not a feature — it’s a foundation.
  • People don’t just forget information — they forget meaning.
  • Base44 is surprisingly powerful for real-world, human-centric compression.

What’s next for AURA

  • Contact-based emotional memory mapping
  • Shared memory capsules between friends or couples
  • API for integrating AURA into other apps
  • Hardware NFC tags for physical memory objects
  • Offline-first QR-based emotional bookmarking

AURA is not just a tool — it’s a new form of digital memory.

Built With

  • android
  • base44
  • local
  • local-nlp-inference
  • on-device-speech/emotion-analysis
  • python
  • qr-encoding
  • secure
  • sqlite
  • storage
  • tensorflow-lite
Share this project:

Updates