GLYSSN

I wanted to know what it would feel like to actually look at a place instead of just walking through it. Glyssn points a phone at a scene and turns what's there into sound: color sets the key, brightness sets the tempo, motion drives the rhythm, saturation wets the reverb. Hold still on a moment and you get a short piece composed from exactly what was in front of the lens. Keep moving and it keeps writing, following you.

It's a kind of synesthesia, borrowed and built. The idea was never "make an app that plays music." It was closer to: what if paying attention to a room, really looking at its light and its color and what's moving through it, could be handed back to you as something you hear. I wanted to find out if a person could listen their way into noticing a place they'd normally walk past. And past that, whether it could hold up live, in the hand, in real rooms, not just as a proof of concept but as something that actually works while you're standing somewhere.

Photo mode captures a single moment into a finished piece. Stream mode never stops, composing continuously as you move. Sessions are saved so what you made doesn't disappear the moment you close the app. I directed all of it, the mapping logic, the architecture, what a "finished" piece should even sound like, working with AI to move fast on the build itself.

It's live on TestFlight, and I have video of it running along with real product shots.

Where it's headed: right now it reads a single scene. The bigger version scans an entire room and lets the room become both the instrument and the level at once, a space you can actually play. That's not built yet, but it's the direction the current version was always pointing toward.

There's also a Moog simulator companion in the works — a Bluetooth-connected synth that pairs with the app, so what Glyssn composes from the camera can be played and shaped on something that feels like hardware, not just on screen.