An accessible PDF needs underlying structure that software and assistive technologies can interpret. Gutenburg’s accessibility model is designed around generating that structure with the PDF instead of relying on a visual-only file followed by manual remediation.
Structure is different from appearance
A document can look correct while still lacking the technical information required for accessible interpretation. Gutenburg’s custom engine is designed to control the generated PDF structure directly.
Semantic content
Gutenburg’s accessibility feature set includes semantic tagging for meaningful document structure, text alternatives for visual information, language metadata, and accessibility-aware table and form workflows. The exact authoring controls used depend on the content being generated.
Author input remains important
The engine can create technical structure, but the author is still responsible for meaning-dependent information and visual choices. For example, the author supplies meaningful image alternative text and controls visual contrast.