Solaris: Exploring musicking with Ai through practice and code

In this performance-lecture, Craig Vear will demonstrate and explain his practice-based research project Solaris. This Digital Score (Vear 2019) is a jazz quartet created from 3 creative-AI agents (piano, bass and TR909 drum machine) and a single human musician (drum kit). It uses abstract visual notation and the behaviours of creative AI to stimulate relationships inside music that are used to communicate musical ideas to the human. The creative-AI is built using an approach called Embodied Intelligence (Vear 2022) and exemplifies the application of the Human-AI Musicking Framework (Vear et al 2023). Solaris uses Creative AI datasets and algorithms to make music from inside the flow (Vear et al 2024). Embodied Intelligence differs from other music-AI approaches insofar that it is not using AI to construct the physical phenomena of music (sound wave), nor the meta-workings of music composition (sequencing of notes). Instead, the AI is powering the realtime generation of core impetus from within the dynamic flow of live music-making (Poltronieri & Vear 2024, Small 1998). Overall, this AI digital score has been designed to make humans more creative. Tree Mountains LIVE DEMO | Solaris (bandcamp.com)

Craig Vear 

Craig Vear is Professor of Music and Computer Science at the University of Nottingham split between music and the mixed reality lab. His research is naturally hybrid as he draws together the fields of music, digital performance, creative technologies, Artificial Intelligence, creativity, gaming, mixed reality and robotics. He has been engaged in practice-based research with emerging technologies for nearly three decades, and was editor for The Routledge International Handbook of Practice-Based Research, published in 2022. His recent monograph The Digital Score: creativity, musicianship and innovation, was published by Routledge in 2019, and he is Series Editor of Springer’s Cultural Computing Series. In 2021 he was awarded a €2Million ERC Consolidator Grant to continue to develop his Digital Score research. 

The Digital Score (digiscore.github.io)