Participants 2025

DrilX

AI Song Contest 2025 / participants

TEAM / DrilX
SONG / Some Might Say

About the TEAM

Both Swiss and San Francisco Bay Area native, Tristan Zand is a conceptual intermedia artist / musician / developer. He gravitates between the Bay Area, Iceland, and Switzerland.

Active performing and recording artist since the mid eighties, his music ranges from experimental jazz core to generative / procedural electronic music. After studying electro-acoustics he never stopped working on various musical and intermedia projects integrating performative setups in an attempt to join both analog and digital worlds together.

His past and current professional experience in network-based, device software app development, but also in science and medical informatics compose his technical expertise, and strongly shape his artistic creativity, including in his understanding and use of artificial intelligence related technologies.

Memories and their incomplete sustainability is a constant focus of his work, both through the conception of art and technology protocols and the seamless use of AI models to that goal.

About the SONG

“Hide the VHS” is the tenth song of an album I composed primarily with generative AI.

It is part of my ongoing intermedia research into how human creators can meaningfully collaborate with AI across all layers of artistic expression.

Created during what I view as a crucial, and uncanny, early phase in the emergence of these tools, the album is curated as a cohesive, though abstract, musical narrative. It reflects the distinct creative voice of one of my musical personas, DrilX.

"

About the HUMAN-AI PROCESS

I used generative AI, primarily Stable Audio Open, for most of the instruments and voices, integrating the curated stems either directly from the prompts or after manual editing with traditional audio tools. My goal was to preserve the essence of my usual collaborative workflow, this time replacing human contributors with their AI counterparts.

I also integrate various sounds and phrases from self-trained models, some based on datasets I began assembling as early as the mid-1990s, using tools such as IRCAM’s RAVE. This adds another more personal and historical dimension to my AI songs in this project.

Check out the other
songs of 2024