NaN is also a lab where creativity meet the latest technology. Discover the self-initiated projects fueling our typographic practice.
Trial fontsMachine Learning Font

Could a machine learning model be trained to generate realistic (or just interesting) letterforms? As it turns out, yes, albeit with mixed results. A styleGAN model was trained in RunwayML on a dataset of 2674 Google fonts organised as individual image-per-glyph in Drawbot. Runway-generated images were then piped via Python in to GlyphsApp to process the final font. Full project and downloadable fonts released under the SIL Open Font License:
Generative Fonts

GenerativeFonts.xyz is a collection of procedurally generated free-to-use, free-to-modify display fonts and the codebase that was used to design them. ‘Procedural’ or ‘generative’ is another way of saying an algorithm was essential to the output and frankly algorithm to me is just another word for the design. Many of these ideas began on paper and were reverse-engineered in to a system with more than a hint of randomness thrown in.
Confusify

A confusable is a glyph that is so similar to another that it may be confused for it. Confusify.py generates fonts that when typed automatically swap confusables for the original. All confusables generated are existing characters in the source font. In that sense it’s a remix.
Select pre-processed fonts released under the SIL Open Font License:
This Font Does Not Exist

In collaboration with Production Type we trained a machine learning model on around 30,000 font bios to generate descriptions of speculative fonts.
Follow us on Instagram
More on Instagram

