What happens when we train RNNs to model fonts as vector (SVG) drawings?

We train a powerful generative model of fonts as SVG instead of pixels. This highly structured format enables manipulation of font styles and style transfer between characters at arbitrary scales!

For more details, visit the source.