FRANK AND FREE and the Origami of AI Sound Production

Piotr “DeSade” Kucharzak exemplifies a fresh approach to music production: here, AI is not a substitute for musical skill but expands the producer’s toolkit. The machine becomes a partner, not an all-in-one solution—a stance that clearly distances itself from pure automation. DeSade emphasizes that AI shines especially where human abilities reach their limits, such as in the complex world of K-pop: arrangements that would be nearly impossible without AI suddenly become manageable, without losing one’s own musical signature.

In conversation with FRANK AND FREE, it becomes clear that this machine ethic requires patience and an ear for the unpredictable. The workflow is marked by a spirit of experimentation, but also by respect for the creative process. DeSade uses AI to create sonic spaces that unfold like origami from tension and noise—a patch geometry where the human sets the direction and the machine provides impulses. Not for workflow fetishists, but definitely a dream for chaos aesthetes.

The fictional K-pop band NEEON DREAM is more than a musical experiment—it is an integral part of the cyberpunk universe “Fabula Ensis,” created by DeSade together with author Lillith Korn. Here, literature and music merge into a carefully crafted total work of art, with AI contributing both narrative and sonic dimensions. The band primarily exists in the novel’s world but is made audible and tangible through AI-driven productions. It’s a clever move that makes the band’s concept itself part of the story.

It gets especially interesting when DeSade describes how AI has, for the first time, made it possible to fully shape his own messages and characters—including song lyrics in Korean and a stylistic coherence that previously hit practical limits. The result is a band fighting evil and creating sonic spaces where liberation and identity are central motifs. Context and background knowledge become the acoustic glue for the soul: without them, AI music remains soulless; with them, a new form of authenticity emerges.

DeSade’s workflow resembles a modular patch geometry: the starting point is often stories from the world of Fabula Ensis, which are first turned into song lyrics and then, via Suno, into music. But that’s not the end—the AI provides raw material, which is then further processed through classic production steps like arrangement, recording, and editing. There are numerous iterations, and elements are often extracted, revised, and recombined until the result is both technically and artistically convincing.

The variety of tools used is interesting: Suno is the central element, but ODEA, 11 Labs, and specialized voice AIs are also used to smooth out artifacts and make voices sound more organic. Even image and video AIs like Midjourney or Nano Banana are used for visual storytelling. The machine provides impulses, but the human shapes them further until the desired sonic space emerges. To experience the full complexity, you should watch the video—some patch branches and sound examples can only truly be grasped beyond the text.

With Milan Holz, technical precision enters the picture: as a mastering engineer at Green Room Studio, he ensures that AI-generated tracks get the professional polish they need. The challenges are anything but trivial—AI songs tend to have artifacts, unbalanced frequencies, hi-cuts, and phasing issues. Especially the low end and dynamics require targeted post-processing so that the song can ultimately compete with classic productions.

Milan describes how post-processing becomes mandatory: compressors, equalizers, and specialized tools like Soos are used to iron out the typical weaknesses of AI-generated music. Mastering here is like a surgical sound operation, removing disturbing frequencies and adding missing punch. Despite all the technology, the human remains in control—and shows that machine aesthetics and craftsmanship must go hand in hand to achieve professional results.