Analyzing 50k Fonts Using Deep Neural Networks
Erik Bernhardsson scrapes 50,000 fonts and trains neural networks on them, then explores the learned font space with visualizations and generated letterforms.
No account needed · sign in to keep your taste everywhere
More like this
Pages that share this page's themes. Or browse all of Tech.
Hey, Computer, Make Me a Font
A personal journey building a generative ML model from scratch and teaching it to produce real TrueType fonts, with glyp…
font.fish
An explorable map of thousands of fonts arranged by visual similarity using UMAP and t-SNE, so you can swim through type…
Coding Font
A game that pits programming fonts against each other in head-to-head rounds so you can find the coding font you like be…
Avería - The Average Font
The story of a font made by averaging every typeface on a computer, with specimens, process images and free downloads of…
Enfont Terrible
A self-described terrible type foundry presenting deliberately awkward typefaces as a tongue-in-cheek take on font desig…
Science of Typography
A typography tool where you pick a font, size, measure and leading and see how each choice changes a block of text, with…
Stumble onto more like this
One click takes you to a page you didn't know you were looking for. Rate it and the next one lands closer to your taste.
Add to Browser — it's freeWorks in Chrome, Edge and Brave. No account needed.