Three funded
PhD fellowships

This position explores data- and interaction-driven approaches to sound- and music-making with machine learning. The fellow will curate small, context-specific datasets; train, fine-tune, and appropriate generative audio models; and use machine learning to craft embodied mappings between the human performance space (gestural and tactile controllers, movement, multimodal input) and the latent space of neural audio synthesis models, treating their parameters as expressive, performable resources for shaping musical structure in real time.

The fellow will compose and premiere new works with these systems and co-author publications, open datasets, models, and artist-readable documentation.

This position investigates embedded generative audio systems as musical instruments, exploring how machine-learning models for sound generation can be adapted, optimized, and deployed on resource-constrained computing platforms for real-time musical interaction. At the intersection of generative audio, embedded computing, digital lutherie, and research-creation, the fellow will develop instruments integrating neural audio models and custom interaction strategies, addressing questions of computational efficiency, latency, model architecture, and expressive control.

Particular attention will be given to variational autoencoders (VAEs) and related neural audio architectures, as well as to the integration of electronics, sensors, and embedded audio platforms. Through iterative instrument design, composition, and performance, the doctoral project will investigate how the constraints and affordances of embedded machine learning can become compositional and performative resources, contributing new approaches to embodied generative AI for music.

This position focuses on distributed and embodied AI for music, exploring how intelligence and musical agency can be distributed across bodies, musical instruments, and technological devices. This PhD/Doctoral student may investigate combinations of performers and embedded computational agents using physiological or gestural sensing, TinyML, microcontrollers, and one-to-many / many-to-many communication as a means to explore musical systems in which decision-making occurs across interconnected nodes.

At the heart of the position is a question of how distributed AI systems might "think" through material or bodies when computational power is limited, but communication between embodied nodes is rich and complex.

For example: what information is sensed locally? What is communicated between performers or devices? What is the temporality of this communication? Where do computational decisions occur? What can these architectural considerations contribute to the field of music, and conversely, what can musical experimentation contribute to how the architecture of AI systems is conceived?

Intersecting concepts of biofeedback and cybernetics with digital lutherie, human–computer interaction, and research-creation, the fellow will explore creative interfaces for distributed musical systems. Projects might explore physiological or gestural computing, networked microcontrollers, on-device sensing and inference, TinyML, or communication among multiple performers and computational agents. These technologies may be investigated through composition, performance, installation, or new forms of digital musical instrument.

The position particularly welcomes candidates interested in questioning assumptions about where intelligence, agency, and control reside within computational systems, or systems of musical thought. Feminist and queer theory, disability studies, critical approaches to technology, non-Western epistemologies, and other cultural or philosophical traditions may provide not only frameworks for analyzing these systems, but models for designing alternative relationships among bodies, machines, and musical agents.

These orientations are invitations rather than requirements, and candidates are equally encouraged to bring other theoretical, technical, artistic, or practice-based approaches to these questions. I am curious to hear your ideas!