Loading...
Loading...
We build open source tools that help universities find accessibility barriers, apply supported partial fixes, and review the results with people who understand the content.
Every letter in AELIRA represents a commitment.
Accessibility is the foundation on which every Aelira feature is built. We did not start with a technology and go looking for a problem. We started with the reality that millions of students encounter inaccessible course materials every semester, and we worked backward from there. The platform checks supported files against relevant WCAG 2.1 AA and PDF/UA criteria, then produces findings and eligible partial fixes for human review. We work toward content that is perceivable, operable, understandable, and robust actually requires. Accessibility is not a feature we bolted on after launch. It is the reason the company exists.
Twenty-six percent of US adults live with a disability. When a professor uploads an inaccessible PDF, they are not making a formatting choice. They are excluding students from the material those students paid to access. We believe that how content was authored should never determine who can learn from it. Equity is the principle that drove us to make Aelira open-core and to price it for higher education budgets, not enterprise procurement cycles. Accessible content should be the default, not a retroactive afterthought squeezed in before a compliance deadline.
Aelira is purpose-built for higher education. The content types we handle (course syllabi, lecture slides, research papers, exam materials, multimedia recordings) reflect the real artefacts that move through universities every semester. We integrate directly with the systems faculty already use. Canvas has maintainer-verified integration testing; Blackboard remains experimental, while Google and Microsoft connectors need institution-specific acceptance checks. Our commitment extends well beyond any single regulation. Wherever learning happens, the materials should work for everyone.
Compliance is the floor, not the ceiling. Meeting WCAG 2.1 AA is necessary, but true inclusion means designing for the full spectrum of human ability: screen reader users, keyboard-only navigators, people with colour vision deficiency, and learners with cognitive or neurological differences. We built Aelira not to produce documents that pass an automated checker, but to produce documents that actually work for the people who need them. Every remediation should make content more usable, not merely more technically conformant.
Most accessibility tools stop at detection. They generate a report, hand it to someone who is already overworked, and hope for the best. Aelira's key differentiator is that it remediates: faculty upload broken content and receive fixed content back. Our AI-powered engine adds alt text to images, tags document structure, repairs reading order, fixes colour contrast, and generates proper table markup. Detection without remediation is a to-do list. Remediation is the actual work, and that is what we chose to build.
Manual accessibility work takes time away from teaching, research, and students. Aelira can scan supported files and propose or apply eligible partial fixes. Reading order, complex tables, subject-specific alt text, and other judgments still need a knowledgeable person. A pilot should measure the effort saved on your own material.
To help universities make measurable accessibility progress with open tools and affordable support.
Faculty need practical help with complex course materials, including LaTeX. Aelira Core is beta software, with documented limits and human review requirements. The open source code and paid support plans let institutions assess it against their own content and budget.
Because accessibility tools should be free, transparent, and community-driven.
We believe universities shouldn't pay high enterprise prices for closed-source software. Our open-core platform at github.com/Aelira-AI/aelira-core (the CLI is MIT-licensed, the engine is AGPL-3.0) is free to download, install, and modify, with nothing feature-gated. We make money by selling support, hosting, and professional services, not by locking you into expensive licenses. This aligns our incentives: we succeed when you succeed.
Under the US DOJ Title II web rule, public entities with jurisdiction populations of at least 50,000 generally have until April 26, 2027; smaller public entities and special districts generally have until April 26, 2028, subject to the rule's exceptions. Read the DOJ rule.
Faculty need accessible course materials and a way to review changes before those materials reach students.
The cost of inaction: Inaccessible material can prevent students from using essential content and leave institutions with unresolved legal obligations.
Existing tools fail faculty:
Aelira offers scans, supported partial fixes, LaTeX workflows, and review artifacts. An institutional pilot should establish fit and effort before rollout.
Aelira can propose or apply supported partial fixes and report what remains open. Subject-matter and accessibility review remain necessary; generated output is not proof of conformance.
Supported LaTeX workflows can produce MathML and review artifacts. Faculty and accessibility specialists should test the output, especially complex packages and equations.
The CLI supports batch processing of eligible files. Throughput, output quality, and review effort vary with the documents; test representative samples first.
A fully local configuration can keep student content on institution-controlled infrastructure and support FERPA and residency review. Cloud providers must be disabled and the deployed egress, logs, cache, scratch, and backup paths verified.
100% free to download, install, and modify. Run it commercially, distribute copies, and keep modifications private for internal use. We sell support and managed infrastructure, not software licences.
Supported workflows provide findings and eligible partial changes. Some inputs need manual work or cannot return a remediated file. Review actual output for content preservation, structure and use with the student's assistive technology; involve accessibility or subject expertise where needed.
We believe accessibility tools should be free and transparent. Every university deserves access to compliance tools, regardless of budget. That's why Aelira is open core: the CLI is MIT licensed, the engine is AGPL-3.0.
We chose our open core model intentionally: MIT on the CLI for maximum flexibility, AGPL-3.0 on the engine to ensure the platform stays open. Universities can run Aelira for free, modify it for their needs, and contribute improvements back to the community. We make money by helping you succeed, not by locking you in.
Self-hosted AI means you control your data. Open source means you can audit our code. No black boxes, no vendor lock-in, no data leaving your servers without your permission.
We focus on LaTeX accessibility, PDF review workflows, and practical evidence for faculty. We publish what the product can do today and where people still need to intervene.
Aelira is developed and maintained by Aelira AI Pty Ltd, a technology company specializing in AI-powered accessibility solutions and open source software development.
Founder RD (Reg) Crampton maintains the project. You can review his public work on GitHub and LinkedIn. We publish product limits so institutions can evaluate the software on their own terms.
A practical way to assess scope, maturity, and your own workload.