Using artificial-intelligence tools to make LaTeX content accessible to blind readers

TL;DR

This paper investigates using AI tools to convert LaTeX documents into accessible HTML for blind users, addressing the challenge of inaccessible STEM content in digital formats.

Contribution

It demonstrates that large language models can generate highly accessible HTML from LaTeX, improving accessibility for blind readers in STEM disciplines.

Findings

Generated HTML is highly accessible for blind users.

AI interpretation can sometimes alter content meaning.

Potential for standardizing accessible STEM content.

Abstract

Screen-reader software enables blind users to access large segments of electronic content, particularly if accessibility standards are followed. Unfortunately, this is not true for much of the content written in physics, mathematics, and other STEM-disciplines, due to the strong reliance on mathematical symbols and expressions, which screen-reader software generally fails to process correctly. A large portion of such content is based on source documents written in LaTeX, which are rendered to PDF or HTML for online distribution. Unfortunately, the resulting PDF documents are essentially inaccessible, and the HTML documents greatly vary in accessibility, since their rendering using standard tools is cumbersome at best. The paper explores the possibility of generating standards-compliant, accessible HTML from LaTeX sources using Large Language Models. It is found that the resulting documents are highly accessible, with possible complications occurring when the artificial intelligence tool starts to interpret the content.