Project Info
Table of Contents
Spatial Audio Weapons App
A whimsical multiplayer playground where kids trigger sound-based "weapons" and hear spatialized effects that change in volume and direction based on everyone’s real-time position in the arena.
The Challenge
Our brief was to turn an imaginary battlefield concept into something children could actually hear and feel. We needed spatial audio that reacted instantly to movement, synchronized between four devices, and fit inside a lightweight SwiftUI interface that kids could understand without instructions.
Our Solution
I ran point as project manager and front-end lead while the team co-designed the experience. We built a SwiftUI controller that tracks player position via a MultipeerConnectivity mesh augmented with calibrated iBeacons, feeds coordinates into an AVAudioEngine graph, and applies per-weapon attenuation curves so volume and panning evolve as players move. RealityKit anchors visualize teammates, keeping the sound cues intuitive.
My Role
Project-managed the three-week sprint while designing and coding the SwiftUI interface, AVAudioEngine graph, and the networking layer. I stood up the MultipeerConnectivity session, tuned an iBeacon grid for indoor positioning, and kept all the moving parts synchronized so the experience felt instant for kids sprinting around the arena.
Key Contributions
- •Facilitated sprint rituals, playtest agendas, and daily priorities so the cross-functional squad shipped a cohesive demo in three weeks.
- •Designed and built SwiftUI control surfaces with color, haptics, and responsive layouts so young players could learn each weapon in seconds.
- •Implemented the hybrid MultipeerConnectivity + iBeacon location bus to broadcast player transforms and weapon events under 40ms.
- •Prototyped the AVAudioEngine attenuation curves, balanced gain automation, and folded playtest feedback into rapid UI/audio tweaks.
AI Matching Pipeline
Weapon trigger broadcast
When a player fires, their device broadcasts the weapon type, timestamp, and transform to peers.
Spatial mix computation
Each receiver calculates the relative vector to the firing player and looks up attenuation/panning settings.
Audio playback
AVAudioEngine applies the calculated gain, pan, and reverb profile before playing the weapon sample.
Technical Highlights
- •Custom AVAudioEngine graph with per-weapon mixer nodes and real-time parameter automation.
- •MultipeerConnectivity mesh networking augmented with iBeacon calibration for reliable indoor positioning and low payload updates.
- •RealityKit scene anchors that mirror peer positions to keep the sonic cues grounded in visual context.
- •SwiftUI gesture system that maps child-friendly interactions (tap, drag, hold) to different weapon behaviors.
Challenges & Learnings
Spatial audio loses credibility fast if latency drifts, so we instrumented every hop and trimmed payloads aggressively. The project reinforced how important it is to test with real movement — lab tuning never exposed the doppler-like artifacts we discovered once kids began sprinting.
Future Improvements
- •Add adaptive EQ so weapon sounds automatically duck under facilitator voice instructions.
- •Introduce procedural sound layers that react to ambient noise levels in the play space.
- •Ship a teacher dashboard for activity presets, safety zones, and participation tracking.
Tech Stack
frontend
SwiftUI interface with custom haptics
audio
AVAudioEngine, spatial mixer nodes, Core Audio
networking
MultipeerConnectivity peer mesh with iBeacon positioning for real-time sync
three D
RealityKit anchors and simple particle effects
tooling
In-app telemetry overlay for latency and gain monitoring
Team
4 team members:
- •Project manager & front-end/audio lead (me)
- •Gameplay programmer focusing on movement tracking
- •Experience designer shaping the imaginary battlefield narrative
- •Audio designer crafting and mastering weapon samples