Tools and Technologies for Accessible Math

Overview

This document provides key tools for creating, converting, and testing accessible math—including equation editors, LaTeX/MathML workflows, graphing tools, and screen readers. It also offers guidance on when to use each tool and important considerations to keep in mind.

Working with More Complex Math Editors

When to use more advanced complex math editors in general

• When you are working with advanced math, physics, statistics, or engineering

• You’re working with complex, multi-line equations, matrices, or piecewise functions.

• You need precise control over alignment, spacing, or presentation.

• You’re preparing formal documents, research papers, or STEM content beyond what Canvas’s built-in editor handles well.

• You want to reuse LaTeX code across different systems

When to use a LaTeX editor

• Multi-line proofs or piecewise functions (a function defined by multiple sub-functions, where each sub-function applies to a certain interval of the main function's domain)

• Publication-ready math

When to use a MathType editor

• Multi-line proofs or piecewise functions

• Have money to spend on it

When to use a Desmos editor

• Graphs, interactive math

Things to keep in mind when using more complex equation editors

• The equations generated are professional and of textbook-quality output.

• Learning how to use these types of math editors can be more challenging, especially if you’re not familiar with LaTeX.

• These types of math editors are not as point-and-click friendly

• Some of them require paid subscriptions and some are free

Testing Tools (e.g., NVDA, JAWS, VoiceOver, MathCAT)

The three major screen readers used for testing are for both Windows and Mac Platforms, they include:

Windows Platform

NVDA (NonVisual Desktop Access):

• Free and open source are typical users

• ESpeak/SAPI5 for the voice engine

• Popular among testers and developers

JAWS (Job Access with Speech):

•  Enterprise and education are typical users

• Uses Eloquence/Vocalize for the voice engine

• Most widely used by experienced blind users

Mac/IOS/IPad OS Platform

VoiceOver:

• Built-in voice engine

• Apple users

• Integrated into Apple devices, ideal for mobile and cross device testing

NVDA uses MathCAT to speak math clearly, support multi-language math reading, and let users navigate complex expressions (e.g., zooming into a fraction numerator).

Tools for converting math content (LaTeX ↔ MathML ↔ accessible text)

Here are 9 common tools for converting math content, they are in no particular order

MathJax
Converts: LaTeX → MathML (and the reverse with some extensions)
Features:

• Renders LaTeX beautifully in browsers

• Exposes MathML for screen readers

• Highly used in education platforms and LMS systems

MathType
Converts: LaTeX ↔ MathML
Features:
WYSIWYG editor
Exports to many formats: LaTeX, MathML, EPS, images
Integrates with MS Word, Google Docs

Equatio (Texthelp)
Converts: Handwriting → LaTeX / MathML / spoken text
Features:

• Predictive math input

• Voice-to-math=

• Makes math screen-reader-friendly

• Exports editable math

Desmos (with accessibility features)
Converts: Math expressions → speech text (accessible math)
Features:

• Screen-reader support using text descriptions

• Used heavily in K–12 for accessible graphing

Pandoc
Converts: LaTeX → MathML (via HTML or EPUB)
Features:

• Command-line tool

• Useful for converting documents between scientific formats

• Can embed math in various output formats

KaTeX
Converts: LaTeX → MathML (accessibility output mode)
Features:

• Fast browser-based renderer

• Can expose MathML for AT (assistive tech)

LaTeXML
Converts: LaTeX → MathML (high fidelity)
Features:

• Very accurate semantic conversion

• Used by math-heavy publishing systems (arXiv, NIST, etc.)

WolframAlpha / Mathematica
Converts: Input expressions ↔ MathML / natural language
Features:

• Symbolic interpretation

• Math export as MathML and image formats

SymPy
Converts: LaTeX ↔ MathML (via Python)
Features:

• Good for automation scripts

• Symbolic math engine

Automated vs Manual Accessibility Checkers (and their limits)

Automated accessibility checkers are tools that scan web pages or digital content using algorithms to detect known accessibility issues.  Several examples include WAVE, axe DevTools, lighthouse, site improve and accessibility insights.  Advantages over manual checks include speed and scale, consistency, early detection and integration.  Automated checks can test hundreds of pages quickly, removing the human bias, helping to catch basic errors early in development and can be embedded in CI/CD pipelines (automated workflow that helps developers build, test, and release software more efficiently and reliably) for ongoing monitoring.  Automation accessibility checkers have limitations in the fact they may mislabel elements due to errors in their complex coding; they also can’t evaluate usability such as navigation and cannot check content meaning.     

Manual accessibility checks involve human evaluators, testing computer interfaces for accessibility and usability barriers.  Several examples include keyboard only navigation tests, screen reader compatibility testing with NVDA, JAWS and VoiceOver and a manual review of color contrast, labels, focus states and reading order.  Advantages over automated checkers include detecting real-world usability issues automation can’t, the ability to assess visual, cognitive and interactive accessibility, helps to confirm and prioritize issues and evaluates if content makes sense and is usable.  Manual checks have limitations in the fact that they are slower than automation, need trained accessibility specialists and may vary between testers if guidelines are loosely applied.     
It is best practice to combine both, using the quick scan and baseline compliance checks of the automation and the usability and content validation of the manual.

Recommended Resources

For more information, see Recommended Resources: Tools and Technologies for Accessible Math.