Bonsai: Wiring the World into Sound and Image

Bonsai (https://bonsai-rx.org/) is an open-source visual reactive programming language that lets people compose streams of data into real-time interactive systems. Though it was born in the neuroscience lab, its dataflow model is ideally suited to creative computing: every input is a stream of events, and every transformation is a composable, real-time operator. Bonsai has a rapidly growing ecosystem of packages for interacting with hardware, creating computer vision workflows, and running live machine learning models. The language can also be easily extended with packages and plugins. This session introduces Bonsai through the lens of an interactive installation. Starting from a blank canvas, we will build a live demo that reads from a camera and microphone, processes those signals in real time, and routes the results to visual and audio effects. After the demonstration, attendees will have the chance to fork the demo and experiment: swapping sensors, adjusting transformations, or remapping input/output relationships. No prior experience with Bonsai is required. Participants will leave with a working Bonsai workflow, an understanding of its core programming model, and a template they can extend for their own creative projects. 

Workshop requirements: Audience level is beginner (programming curiosity sufficient; no Bonsai experience required)

About NeuroGEARS:

 This demo will be facilitated by NeuroGEARS (https://neurogears.org/), an independent research group and company developing the open-source Bonsai programming language. Bonsai began as a laboratory instrument, developed by Goncalo Lopes (now director of NeuroGEARS) during his PhD as a way for neuroscientists to integrate and synchronise diverse data streams from cameras, peripheral sensors, and electrophysiology equipment without writing low-level code. Over the past decade it has grown into a mature, community-driven platform used in research worldwide and, increasingly, artists for creating interactive art. The language's reactive model also has a home beyond the lab; it provides a software framework for live performance art, audiovisual installations, and anything demanding routing and transforming sensor data in real time. This demo is part of a broader effort to introduce Bonsai to the creative computing community and to learn from artists and makers who may push the tool in directions the scientific community never imagined. Individual facilitators: Andrew Erskine (https://www.andrewerskine.uk/) has over a decade of experience working in neuroscience labs as a researcher and software engineer. He then moved to NeuroGEARS where he has helped develop the Bonsai ecosystem and build the community through school outreach, conferences and workshops. Nicholas Guilbeault (https://ncguilbeault.com) is a research software engineer and neuroscientist. During his PhD he developed real-time, closed-loop fish VR experiments combining computer vision, behavioral tracking, and visual stimulation. At UCL's GCNU he developed Bonsai.ML, bringing machine learning into the Bonsai ecosystem. He has contributed numerous packages and applications to the Bonsai community.