Network Music Festival

Sound without Borders // 15-18th July 2020

Go to Online Exhibition

Using Machine Learning to Build Musical Instruments in the Browser // Louis McCallum and Mick Grierson

Thursday 16th July 1400-1700 (GMT +1)

Interactive Machine Learning (IML) is a great approach for building interactive systems, including new music instruments and their mappings from input to sound. Building browser based tools is useful because Javascript is a fast growing language with low barriers to entry, additionally, browser based tools require little of the installation and dependencies that plague other data science endeavours.

In this workshop, we teach participants to use machine learning to map an input (video, audio, motion…) to an output (MaxiSynth and MaxiSampler) and develop a new musical instrument all within the MIMIC browser based platform. We plan a simple tutorial of machine learning concepts, followed by an introduction to the website and tools themselves, followed by an open workshop where participants develop their own classifiers and mappings under the guidance of workshop leaders.

Level: intermediate

15 max. participants

Sign up here:

Louis McCallum is an experienced software developer, researcher, artist and musician. Currently, he is an associate lecturer and researcher at the Embodied AudioVisual Interaction Group, Goldsmiths, University of London. Recent work has involved developing the IML tool Wekinator with Dr Rebecca Fiebrink, and also as a researcher on the AHRC funded MIMIC project ( building web based tools for musicians to use machine learning. Currently teaching a module on machine learning for creative practitioners, he is excellently placed to provide insight into how machine learning can be used as a design tool for musicians, as well as the ways state of the art machine learning techniques can be incorporated into musical practice.

Mick Grierson is Research Leader at UAL Creative Computing Institute (CCI). His research explores new approaches to the creation of sounds, images, video and interactions through signal processing, machine learning and information retrieval techniques. He is Principal Investigator on the £1million Machine Learning and Artificial Intelligence project MIMIC, which is a partnership between CCI, Goldsmiths, Durham University, Sussex University, and Google’s Project Magenta. He is also lead academic researcher on the £1million Wellcome Hub project “Heart and Soul at the Hub”, exploring technology assisted research methods with learning disabled and autistic people, disabled arts charity Heart n Soul, RCA and UCL. Hardware and software based on his research has been widely used by world leading production companies, tech start-ups and artists including the BBC, Channel 4, Sigur Ros, Christian Marclay, Martin Creed, Arca and many others. Mick is also the director of the Daphne Oram Collection, and co-founder of the Daphne Oram Trust.