There are compelling opportunities for empowering the live coding
community with new artistic processes and outcomes that may arise from
the integration of real-time interactive signal processing and machine
learning technologies, in a scalable and accessible environment such
as the web browser. This workshop is part of the MIMIC research project,
which explores how machine learning and machine listening can be
designed in a user-friendly way and provide for live coding, rapid
prototyping and fast development cycles of musical applications.
Participants will get familiar with our live coding language design
environment and language design techniques, preparing them to work in
the creation of their own live coding languages. Participants will
learn how to express sound and music concepts using signal processing
expressions. They will experiment with selected machine learning models
and machine listening JavaScript classes and explore some of their
direct benefits, including beat detection, pattern detection and
generation.
Participants will gain understanding about workflows in our system that
will allow then to develop their own live coding mini-language using
language design techniques. For instance, they will learn how to
create, inspect and change language specifications to generate language
parsers. Participants will also explore practical techniques for
performing with the live coding mini-languages they have constructed
or customised, including live coding with interactive machine learning
and pre-trained generative models.
Additionally, the workshop will include a group discussion about
participants’ experience learning and using our system. As a result of
this workshop, we believe participants will gain a better understanding
of what signal processing, machine learning machine listening are in
the context of live coding language design and performance.