28K820
I build fictional characters and write their songs. Real lyrics with music like soundtracks to a life that could've been. Started because I wanted collaborators but was working solo across Europe. Turns out AI music generation gets interesting when you stop treating it like a tool and start treating it like a band member with terrible taste that needs coaching. My approach: give each character actual depth - not just a vibe, but a whole identity. Characters like Drift Evans, Sad Bill Smiles, IRL, Jone Minus One, and others document life from different perspectives, places, locations, and experience. I'm not afraid to mess with Manchester post-punk archaeology or fuse traditions that shouldn't work together (but do). The lyrics matter because the characters matter. Technical side: I've built reusable methods for coaxing novel sounds out of Suno - controlled chaos punctuation, acoustic-to-electronic processing, geographic scene evolution, stutter effects that actually serve the song. The goal isn't complexity for its own sake. It's finding sounds that match what the character would actually make. Human imperfection over AI defaults.
