AI Sonic Explorations

This course is a survey and a series of hands-on
workshops exploring the critical, speculative, and
also practical ways of applying Artificial
Intelligence to working with sound and musical
composition. In particular, we will be focusing
on using Machine Learning models that learn from
input data to optimize towards a certain task.

In this class, we will look at the uses of ML as a
way to deepen our relationship with sound and
music production. Particularly we will look at
machine learning techniques that focus on machine
and human collaboration rather than ML models as
solely an open "toolbox" for new sound techniques.
This class will learn how to organize and work
with large sound datasets, generate new raw audio
with neural networks, work with existing plug-ins
that have machine learning capabilities, as well
as generate midi and compositional structure by
training our own AI models. It will be a series of
hands-on workshops, critical discussions, and
investigations of generative sound techniques
where we will build our own projects (whether
sound art installations, musical compositions,
electronic music production, or interfacing
Machine Learning sound models on the web).

While this isn't a programming course, experience
with sound production techniques, digital audio
workstations, and a basic understanding of
programming concepts are required, as we will need
to be able to go into the code and steer the
models toward artistic goals. There will be a
series of technical mini-projects as well as a
final project of the students own choosing.

Requirements:

Some basic knowledge of sound production/ audio
concepts.
A basic grasp of programming concepts: variables,
strings, functions, object oriented programming:
Intro To Programming or some kind of programming
equivalent.