Machine Learning and Type Design 01: “Machine Learning and Type Design”

*This article is created based on a presentation given at ATypI 2019 Tokyo.

The term “artificial intelligence” or “AI” is becoming increasingly common. A method used for this technology is called machine learning. Currently, research is being conducted worldwide in order to utilize this machine learning in type design.

There are two cases in which machine learning can be utilized in the process of type design: typeface design and quality assurance. As 23,060 characters are registered in the Japanese typeface character set Adobe-Japan 1-7, it takes an enormous cost to design them. Therefore, a method to reduce this burden by using machine learning has been proposed. From the perspective of quality assurance, it is believed that machine learning can be used for checking to maintain consistent design for all characters.


Series archive Other / Machine Learning and Type Design