Kandi

Concert chaos, controlled

Comment

Inspiration

We’re both avid concert and festival goers, and after years of attending shows, we realized we kept running into the same two problems every single time: protecting our hearing and losing our friends in the crowd.

At concerts and festivals, sound levels often exceed safe exposure limits — but most people don’t know how loud it actually is in the moment. That’s led to the classic post-show ringing ears after spending too long near the rails.

The second problem is just as frustrating: in massive crowds, groups get split up constantly. Totems help a little, but at large venues they often aren’t enough. There’s no easy real-time way to know where your friends are once you’re separated.

That shared experience inspired us to build KANDI — Kinetic Audio Navigation and Decibel Intelligence — smart concert/festival assistant glasses that solve both problems at once. The name “Kandi” is inspired by the handmade beaded bracelets exchanged at raves and festivals, symbolizing connection, community, and looking out for one another. That’s exactly what we wanted this project to represent: helping people stay safe, stay connected, and fully enjoy live music without losing their hearing — or their friends.

What it does

KANDI tracks decibel levels in real time and calculates safe listening exposure through our Swift app, then streams that intelligence directly to the smart glasses HUD.

With KANDI, users can instantly see:

  • Live dB readings from the audio pipeline
  • Dynamic safe exposure time based on sound intensity
  • Visual warning states when hearing risk thresholds are crossed

KANDI also helps groups stay together in crowded environments using Bluetooth Low Energy (BLE). Each group member’s phone/glasses setup acts like a BLE beacon node, and KANDI estimates relative proximity using signal strength (RSSI) and directional context. The glasses then display intuitive guidance cues (example: “Sarah → left, close”).

So instead of guessing where your friends are — or whether your hearing is at risk — KANDI gives you real-time, wearable intelligence for both.

How we built it

We built KANDI as an integrated wearable system: smart glasses + full iOS Swift app + BLE networking layer.

Hardware + wearable layer

  • Smart glasses provide the in-view HUD for real-time overlays
  • The glasses act as the user’s glance interface so they don’t need to pull out their phone mid-set
  • Overlay states are intentionally minimal: decibel level, safe-time remaining, and friend direction cues

Entire Swift app stack (iOS)

We built the full companion app in Swift to handle sensing, networking, and UI:

  • SwiftUI for the mobile interface and live state updates
  • CoreBluetooth for BLE advertising/scanning, group discovery, and beacon-style proximity tracking
  • AVAudioEngine / iOS audio APIs for ambient sound capture and decibel estimation
  • Exposure modeling logic that converts sound intensity into safe listening duration windows
  • Group session management for friend linking, presence, and reconnect behavior

Bluetooth friend-finding architecture

  • Every user in a group participates in a BLE mesh-like discovery model
  • The app continuously samples RSSI from nearby peers to estimate relative closeness
  • Directional and relative-location hints are generated and pushed to the glasses HUD
  • Designed to remain useful in dense, high-noise environments where visual totems fail

Real-time hearing intelligence pipeline

  • Ambient sound data is processed into a rolling decibel estimate
  • Risk levels are calculated continuously rather than in static snapshots
  • Safe exposure time is recalculated live and surfaced directly on-glasses
  • Warning thresholds are made highly visible so users can act before damage occurs

Our guiding principle was low-friction UX: no menu-diving, no complicated flow, just immediate safety + navigation feedback in the moment it matters.

Challenges we ran into

One of the biggest challenges was refining the core idea into something that was both technically feasible and actually useful in real festival conditions.

Other major challenges included:

  • Designing a glasses HUD that is readable in dynamic festival lighting without becoming distracting
  • Turning raw microphone/audio signal data into stable, meaningful dB and exposure outputs
  • Handling BLE signal fluctuation and RSSI noise in dense crowds full of interference
  • Building real-time friend guidance that feels intuitive despite constantly changing relative positions
  • Balancing technical depth with a dead-simple UX so users can act in seconds, not minutes

In short: building something “smart” is easy to say, but building something that works in loud, crowded, real-world environments is a different beast.

Accomplishments we’re proud of

We’re proud that we built a project that addresses two real, personal pain points we’ve experienced over and over.

What we’re most excited about:

  • Building a complete end-to-end prototype: smart glasses display + full Swift iOS app + BLE coordination
  • Creating real-time hearing protection intelligence instead of passive post-event awareness
  • Implementing beacon-style Bluetooth friend-finding for crowded, real-world environments
  • Translating complex technical data into instantly understandable wearable UI cues
  • Combining safety + social connection in one product rooted in rave/festival culture

KANDI feels meaningful to us because it’s not just a gadget idea — it’s a tool we genuinely wish we had at past events.

What we learned

This project taught us a lot about both product design and human behavior in live-event environments.

We learned that:

  • People are much more likely to protect their hearing when feedback is immediate and visible
  • In high-noise, high-density spaces, UX must be extremely clear and minimal
  • Solving one niche problem is good — solving two connected problems can create a much stronger product
  • Building for real users means designing for stress, distraction, and imperfect conditions

Most importantly, we learned how to turn a shared personal frustration into a solution that could help a lot of other people have safer, better experiences at the events they love.

Share this project:

Updates