Scanned text needs proof, not confident-looking tags.
OCR can make a page searchable while still misreading a name, date, amount, heading, or column order. The strict auto-tagging service therefore rejects scans and PDFs with OCR-derived text before it renders pages for AI verification or generates any tag structure.
Free limits: 5 documents per day per network address, up to 20 pages and 10 MB per PDF. Use the upload below to confirm the routing decision. A rejected scan needs reviewed OCR, source comparison, and assistive-technology checks; it is not converted automatically. Sign up for the review workspace, specialist routing, and hosted HTML.
Up to 20 pages and 10 MB. Processed in memory, never stored.
Read this before you rely on the result
Why a person is still part of accessibility.
What the machine verified
This check inventories every page for authored text, raster or scan content, sparse text that needs recognition, and known OCR provenance. A scan or OCR-derived layer is refused before AI visual verification and before any PDF tag is written.
What only a person can decide
- Whether recognized names, dates, amounts, symbols, and punctuation match the visible source.
- Whether columns, headings, lists, notes, and page furniture were recovered in the intended reading order.
- Which language, handwriting, specialist notation, and image purpose the source actually conveys.
- Whether the repaired result works through real keyboard, zoom/reflow, and screen-reader tasks.
Where this output stands
This result is an eligibility and routing decision only. A rejected scan is unchanged and still needs reviewed OCR or specialist remediation before accessibility testing can begin. This is the industry consensus, not our caveat: the Matterhorn Protocol that defines PDF/UA testing splits its checks into machine-verifiable and human-judgment conditions, the standard checkers pair automation with a required visual check, and the U.S. Department of Justice, extending the ADA Title II deadlines in 2026, observed that generative AI does not yet reliably automate remediation at scale. No tool that claims otherwise is being straight with you, and automated output here never carries a PDF/UA conformance claim.