ASCII art generator with Fable 5.2 (Experiment)
retained text
Saw a video of a flower growing out of glowing characters and wanted a tool that does that to any picture. So I had Claude Code build it. **What it does now:** * drop an image, GIF or video, or just type a word * 9 looks (terminal, halftone, newsprint, matrix…) and your own character sets * turns the picture into a 3D object you can spin. It finds the subject by itself and keeps the background flat. There’s also a depth model running in the browser for real depth * writes a soundtrack from the picture: colour picks the key, detail picks the tempo * exports PNG, GIF and video, with transparent background if you want It’s one HTML file, no backend, nothing gets uploaded. WebGL2 + Web Audio, no frameworks. Started as one prompt, then a handful of short follow-ups like “add templates” and “long images get cropped, fix it”. Try it: [https://winchxyz.github.io/glyph-bloom/](https://winchxyz.github.io/glyph-bloom/) Code: [https://github.com/winchxyz/glyph-bloom](https://github.com/winchxyz/glyph-bloom)
document fetched 2026-11-11 00:40 · canonical https://reddit.com/r/ClaudeCode/comments/1xt1t87/ascii_art_generator_with_fable_52_experiment