From Text to Sound

Beth Coleman and Maurice Jones

00:00

Sold out (not in passes)

Artist-researcher Beth Coleman opens up the first of three Wilding AI skill-sharing sessions by unpacking her appeal for imagining wild AIs »that can be free.« Coleman will also introduce how the Wilding AI collective aims to operationalise this notion of freedom through its collective research-creation practice.

Maurice Jones will then dive into the world of Large Language Models (LLMs) and explore their application in artistic practices. This workshop begins with an introduction to the fundamentals of LLMs, covering their general structure, word prediction mechanisms, prompt engineering, retrieval-augmented generation, and fine-tuning techniques. The session will showcase open-source tools like Open WebUI and examine compelling case studies where LLMs have been employed innovatively in the arts.

In the second half of his workshop, Jones will transition to the realm of multi-agent systems and how the Wilding AI collective mobilises these to compose immersive experiences in spatial audio environments, pushing the boundaries of sound and space.

Event Access

Note: this event is not included in festival passes. Please consult our Accessibility section for an overview of access information.

Presented in collaboration with »Wilding AI,« a research-creation collective led by Beth Coleman, Romi Morrison and Maurice Jones, the »Wilding AI Lab« at CTM 2025 is supported by MONOM Studios, and CTM Festival.

The ongoing »Wilding AI« initiative is supported by MUTEK, MUTEK Mexico, MONOM Studios, 4DSOUND, Neutone Inc., Concordia University, and the Society for Arts and Technology.

Wilding AI is made possible by round 14 of the Goethe-Institut International Coproduction Fund.