Benoît Bovy is a freelance scientific software developer and musician. He has a background in geoscience (geography, geology) and is a regular contributor to the Python scientific / PyData stack (Xarray). He is also active in the Belgian music band Roscoe.
Jupyter already has a rich ecosystem of widgets that together allow using it as a powerful platform for interactive data visualization. However, to my knowledge no generic widget library has been proposed yet for exploring data through sound. In parallel, there exists a few programming environments for interactive creation of sound and music (e.g., Sonic-Pi, FoxDot), but those are isolated applications focused on code and hardly reusable in general-purpose environments.
Ipytone is a widget library that fills this gap by providing many audio components (i.e., oscillators, filters, effects, synthesizers, samplers, etc.) to Python and Jupyter. Those components are part of the Tone.js library, which is built on top of the Web Audio API for creating interactive music and sounds in the browser. Ipytone exposes each component as a widget, making the library very flexible and tightly integrated with the rest of the Jupyter widget ecosystem. Ipytone aims at turning Jupyter into a versatile DAW (Digital Audio Workstation) and at democratizing “data sonification”, a fascinating although still largely unexplored area!
This talk will introduce the audience to Ipytone through various examples and demos, hopefully with live sound! These will range from basic usage (e.g., creating a simple synthesizer and playing it) to more advanced usage (e.g., reproducing NASA’s sonification of a Hubble deep space image in a notebook using Ipytone and other Python/widget libraries).
- Ipytone repository: https://github.com/benbovy/ipytone
- Ipytone examples repository: https://github.com/benbovy/ipytone-examples
- NASA Hubble data sonification: https://www.youtube.com/watch?v=H-Ci_YwfH04
Audience: intermediate (basic knowledge of ipywidgets)