A sensory-blending art & engineering installation exploring how we can blend auditory and visual senses. We designed interactive sound mixing podiums where audiences could alter music in real time while generating evolving projected artwork. This project integrated dynamic lighting, engraved tactile surfaces, and multi-station interaction to create an immersive experience inspired by synesthetic perception.
View More →
This is a bit self-explanatory, I built high-fidelity speakers from scratch including using enclosure modeling, analog crossover circuit design, and driver selection. I also performed hands-on assembly with soldering, wiring, and acoustic testing using signal generators and oscilloscopes.
I built a neural network model to generate new audio clips based on a single audio input, using an unconditional generative architecture. In this model, I implemented a diffusion model that would train on images of music through spectrograms. I trained the model only on one piece of music at a time, allowing for little to no extra data required - due to this, it can be done on any GPU. See "Learn More" for in-depth analysis.
Learn More →