CRANE lab

4 & 5 juillet @ CRANE lab https://linktr.ee/cranelab

https://manolisekmektsoglousite.wordpress.com

In my recent work TrajSynth, created for 32-channel diffusion, I explore the use of artificial intelligence as a generative tool – not for producing entire sound materials directly, but for sonifying the dynamic numerical processes that AI methods inherently generate. These constantly evolving numbers, often ignored in mainstream AI-music platforms, carry a kind of latent energy that I find highly effective for sound creation.

Rather than relying on ready-made platforms that tend to produce superficial and derivative results, I focus on tapping into the internal workings of AI – extracting numerical data from within neural networks or training loops – and translating that into meaningful sonic gestures. This approach has proven more musically satisfying and opens up a broader expressive potential.

What makes this particularly interesting in a multichannel diffusion context is the scale : AI methods inherently generate massive amounts of data, and this aligns beautifully with the spatial complexity that a 32-channel diffusion system can support. In a way, the richness of the AI’s internal processes finds a natural counterpart in the spatial granularity of high-channel-count systems.

That said, technical limitations remain : for example, assigning one neuron to each channel would be an ideal mapping strategy, but it’s currently infeasible due to the immense computational demands. Nevertheless, the creative possibilities of using AI-generated trajectories as spatial and sonic material are vast – and I believe this is only the beginning.