Live coding brings together music and code in live performance. Although writing lines of code in real time is engaging, it can also become too slow at performance time. When using large crowdsourced databases, there is a level of unpredictability of the search results. This situation can bring sounds that are very far from suitable in a given musical context.
MIRLCa is a virtual agent (VA) developed as part of the project “MIRLCAuto: A Virtual Agent for Music Information Retrieval in Live Coding’, which is funded by the EPSRC HDI Network Plus – Art, Music, and Culture theme. This project aims at creatively exploring the use of large collections of sound data in live coding performance through the use of machine learning and music information retrieval algorithms and the use of a virtual assistant that can complement a human live coder in her/his practice.
MIRLCa is a self-built SuperCollider extension and a follow-up of the also self-built SuperCollider extension MIRLC, which is a user-friendly live coding environment that allows the live coder to query crowdsourced sounds from the Freesound online database using MIR techniques resulting in a sound-based music style. MIRLCa includes machine learning algorithms enabling the live coders to design a customisable VA to assist them in their practice. Based on a set of examples provided by the live coder, the system filters the crowdsourced sounds retrieved from Freesound at performance time by only retrieving sounds predicted as “good” sounds. In a second phase, the system will also learn from the code structure of the live coder and respond to the live coder actions.
Dr. Anna Xambó is a Senior Lecturer in Music and Audio Technology at De Montfort University, a member of the Music, Technology and Innovation – Institute of Sonic Creativity (MTI2), and the principal investigator of the EPSRC HDI Network Plus funded project “MIRLCAuto: A Virtual Agent for Music Information Retrieval in Live Coding”. She specialised in human-computer interaction and music technology at Universitat Pompeu Fabra, and completed her PhD in music computing at The Open University. Her passion for sound and music computing kept being nurtured as a postdoctoral fellow at Georgia Tech, a postdoctoral research assistant at Queen Mary University of London, and an Associate Professor in Music Technology at the Norwegian University of Science and Technology. Her research and practice focus on sound and music computing systems looking at novel approaches to collaborative, participatory, and live coding experiences. Her solo and group performances have been presented internationally in Denmark, Germany, Norway, Spain, Sweden, UK and USA.