09-24, 15:10–16:40 (Europe/Amsterdam), Margaret Hamilton @ TNW City
Sonification–using sound to represent data–is a niche technique for exploring complex patterns, expanding the sensory dimensions of data analysis, and discovering musical ideas that are otherwise inaccessible.
In this hands-on session, participants will learn the ins and outs of building sonification pipelines through practical examples with data from healthcare and physics. We’ll also cover key software design considerations for creating flexible and expressive systems that map data into sound. Whether you're a developer, data scientist, researcher, educator, or artist, this session will help you listen to your data.
This tutorial is aimed at people working with data who are familiar with basic components of music, like pitch, melody, harmony, and rhythm, although the ability to play an instrument is not required. The technical part will rely on common libraries in the Python data analysis stack: Numpy, Pandas, Matplotlib, Streamlit, and HuggingFace.
The goal of this tutorial is to introduce you to the challenge of creating generative music that is grounded in, and reflects some underlying data structure–whether temporal, spatial, or statistical–while remaining musically coherent and expressive.
We’ll cover a number of methods, from straightforward rule-based systems to billion-parameter transformer models, highlighting the tradeoffs between control, complexity, and creative potential across different approaches.
During the tutorial you’ll get access to relevant data, and a repository containing end-to-end code examples, including a sonification of healthcare data (ECG signals) and a chaotic physical system (the double pendulum). Using those examples, we’ll explore multiple sonification strategies that can be applied, and multiple aesthetic choices that need to be considered when writing sonification software.
Outline
- A brief history of sound, music, and algorithms. (10m)
- Becoming one with the data – exploring datasets with all your senses. (15m)
- The piano keyboard – absolute minimum of music theory for data scientists. (10m)
- Finding beauty in the dissonance – sonification of chaotic motion of the double pendulum. Step-by-step walk through. (20m)
- Extra beat – sonification of ECG data acquired from patients with various types of arrhythmia. Step-by-step walk through. (20m)
- Team work – exploring ideas together with the audience. (15m)
If you want to experiment with your own sonification ideas during the tutorial (which is encouraged), please bring headphones! 🎧