Blind and visually impaired learners
Need understandable, navigable access to complex subject material that works with their assistive technologies.
About the project
IncluLearn.AI
IncluLearn.AI investigates how visually dominated STEM learning materials can be semantically reconstructed so that they can produce reviewable and different accessible representations.
Starting point
STEM content is more than images with alt text
Diagrams, mathematical expressions, circuits, tables, plots and other technical representations carry subject-matter structure. A short image description is therefore often insufficient to make the same information accessible, navigable and educationally useful.
The project therefore treats accessibility not as a late export problem but as a question of how subject meaning can be captured, reviewed and transformed into suitable representations.
Approach
Semantic first: review meaning before generating representations
The core approach separates subject semantics from individual output formats. A reviewed semantic basis is intended to support multiple accessible representations without repeating the same subject interpretation for every format.
- 1. Understand the source materialIdentify document structure and relevant STEM artefacts.
- 2. Reconstruct semanticsModel objects, relations, values and subject structure explicitly.
- 3. Human reviewMake uncertainty visible, correct it and approve the subject meaning.
- 4. Generate suitable representationsFor example structured text, mathematical representations, audio or tactile outputs, depending on the content and usage context.
User and stakeholder perspectives
Who the project is relevant to
The concrete requirements analysis will continue to be validated during the project. Several groups with different workflows and quality needs are already being considered.
Teachers and material authors
Need to adapt existing materials or create new content accessibly from the outset.
Media centres and conversion services
Bring professional expertise in subject adaptation, quality assurance and multiple target media.
Higher-education accessibility and e-learning services
Are important for institutional workflows, learning platforms, examinations and sustainable integration.
Newsroom
How to follow the project
The four topics separate project development, scientific context, transfer and public activities. Dossiers provide broader syntheses.
Project progress
What is becoming more concrete in the project – from user needs and requirements to prototypes, evaluation, governance and services.
Research radar
Research, standards and methods relevant to accessible STEM learning materials, semantic reconstruction and multimodal AI.
Tools & transfer
Tools, standards, open technologies and transfer questions that can turn research results into usable practice.
Publications & events
Our publications, talks, workshops, events and professional exchange around IncluLearn.AI.
For the advisory board and partners
Open questions are part of the documentation
The newsroom is not intended to present finished results only. Where appropriate, posts also show which assumptions still need validation and where feedback from practice, teaching or accessibility work would be particularly useful.