ΛLEXSTRΛVS ·X

Each release is conceived as a complete narrative work — a journey through landscapes that exist only in sound and imagination.

The musical language draws from European electronic traditions of the 1970s and 1980s — the patient layering of the Berlin School, the elegant melodies of NOVA, the cosmic stillness of Klaus Schulze.

I cannot read sheet music, but I have a clear inner sense of how a piece must sound. My work moves between traditional synthesizer composition, live-recorded instruments, and AI-assisted production — with the human hands carrying more of the weight with each release.

I sketch and compose sound elements on a Yamaha CS-01 — a small analogue synthesizer from 1982 — on the software synthesizer Vaporizer2, and on the Arturia KeyLab Essential controller with Analog Lab Pro, a curated library of classic analog and digital synthesizer emulations. For specific moments, I bring in live-recorded electric guitar on a Yamaha Pacifica, captured through a Focusrite Scarlett 2i2 audio interface — the only human-performed musical layer in an otherwise electronic work. Individual tracks are tested and refined in Audacity before everything comes together in Ableton Live for final production and arrangement. Hours of listening, playing, refining, finding the exact texture.

Beyond the instruments themselves, I write custom Python scripts for audio processing — stem separation, signal analysis, and custom transformations that integrate AI-derived material with my own compositions. This technical layer is part of how a track takes shape.

For specific elements — orchestral accompaniments, string sections, supporting textures — I use Amuse running ACE-Step v1.5 XL Turbo to generate sequences and partial compositions. These are then reworked at the KeyLab Essential, adjusted, refined, and integrated within Ableton Live until they fit the larger work.

Sound design — environmental textures, machine sounds, atmospheric layers — is crafted with the ElevenLabs Sound Generation API through a custom Python pipeline. Where a track calls for a human voice, I work with ElevenLabs Voice Cloning to generate a raw vocal, then process it through voice manipulation tools until the character matches both the music and the visual portrait the narrator carries. The voice that reaches the listener is shaped by my hand. Companion images for each track are created with DALL-E.

Some pieces incorporate real recordings from beyond Earth — sonified data from NASA's Voyager Plasma Wave Subsystem, the actual sound of the interstellar medium captured by the Voyager probes after crossing the heliopause. When the narrative calls for the void to speak, I let the real void speak.

Earlier releases involved MusicGPT as the primary AI-assisted source. Over time, the work has moved toward local tools and — more importantly — toward more of my own hands on the result. With each album, the human share has grown: live instruments, custom scripts, careful production, hours of mixing and mastering in Ableton Live. The AI has become a helpful assistant rather than a co-author.

Each track grows from many directions: from the keys under my fingers, from strings under my fingertips, from sequences I have shaped and reshaped, from real cosmic recordings. I bring them all into a single voice in Ableton Live. The vision, narrative, composition decisions, and final curation are entirely my own. The AI is a tool. The vision is mine.

I am currently working on a trilogy — a quiet homage to the spirit of the 1960s and 70s, to the Pioneer, Mariner and Voyager missions, and to a time when humanity still looked outward in wonder, before turning inward into conflict.