The world around us is vibrating with hidden frequencies, but we only perceive them as static textures. We asked ourselves: "What does rust sound like? What is the frequency of concrete?"
Inspired by the concept of Simulated Synesthesia and the raw aesthetic of industrial hardware, we wanted to build a bridge between the physical and the auditory. DRIFT isn't just an app; it's a "Texture Decomposition Protocol" that translates the visual complexity of the physical world (entropy, roughness, pattern) into sonic textures that you can play and manipulate.
What it does DRIFT is an intelligent synthesizer that uses Multimodal AI to "hear" images.
Capture: You point your camera (or upload a file) at a surface—wood grain, brutalist concrete, a patterned shirt. Analyze: We use Google Gemini 1.5 Flash to analyze the visual texture. It looks for properties like "roughness," "entropy," "luminance," and "pattern regularity." Synthesize: The AI translates these visual traits into audio parameters (oscillator type, harmonic content, distortion, filter cutoff, LFO modulation). Perform: A generative synthesizer engine (Tone.js) instantly patches itself based on the image, creating a unique soundscape that represents that specific material. You can then manipulate the sound in real-time.
Log in or sign up for Devpost to join the conversation.