In this web content, the author shares their journey of learning to build generative ML models from scratch and using those models to teach a computer to create fonts. They introduce their project called ‘FontoGen’ and showcase examples of fonts generated by the model. The author discusses the challenges of generating fonts and describes the model architecture, which is a sequence-to-sequence model trained on text and font embeddings. They explain the training process, the attention mechanism used, and how they achieved good training performance. The author concludes by expressing their excitement about further possibilities and invites feedback from readers.
https://serce.me/posts/02-10-2023-hey-computer-make-me-a-font