audio fingerprints

2019

This project explores the relationship between sound and structure, translating audio data into deterministic, generative artworks. Using Processing IDE, I created a application that listens to a piece of music and transforms it into a unique graphic representation of the song: an audio fingerprint.

The process works by sampling the audio file at one-second intervals, extracting a snapshot of both amplitude and frequency content using Fast Fourier Transform (FFT) analysis. Each moment is then interpreted through a set of visual rules, plotting shapes, lines, and colors that correspond to the spectral characteristics of the sound.

The result is a visual timeline of the song’s tonal content.

Each image in the series represents a different track, and each version of the script applies a different rule set, changing how data points are rendered, connected, or weighted. Some are minimal and geometric; others more chaotic and expressive. But all are deterministic: given the same input and parameters, the system will always return the same visual product.