In a German Bescheid the structure carries the meaning — which box is ticked, which clause sits in which field. As loose text it comes back as fragments.
Reglyph keeps the page as a page and swaps only the printed German. Where the English runs wider than the German it replaces, the type is scaled to its cell instead of pushing the rest of the form out of shape.
This is page 2 of an administrative decision issued by the city of Magdeburg under the Saxony-Anhalt Access to Information Act. Nothing here is a mock-up — drag the handle and compare it line by line.
A German Bescheid packs three different things onto one page: a paragraph of statutory citations where a single altered file number would be a real error, a fee table in euros, and a signature block. Here the signature has been blacked out — and that block is part of the scanned image, not part of the text, so a tool that rebuilds the page from extracted text alone tends to lose it or move it. It comes through in the same place, over the same line.
The first paragraph is a long citation string, and its last line sits tight against the paragraph below it — English is wordier than German and that block had no room left to grow into.


Already have a digital German PDF rather than a scan? German → English document translation
Yes. Save the scan as a PDF — iPhone and Android both have a document scanner built in that does this — and upload it. Image-only PDFs with no text layer are the normal case here, not an edge case.
These are the two characters that decide whether a German scan translates correctly, so they get specific attention: ß is not collapsed to B or ss, and ä/ö/ü are not flattened to a/o/u. Getting them wrong changes the word rather than misspelling it — Maße means dimensions, Masse means mass.
German compounds often translate to something wider than the space available. The English is scaled to the cell it belongs to rather than reflowing the page, so the table and form structure holds its shape — on very tight cells that means the English is set a little smaller than the German around it.
Simple, scan-friendly pricing. Pages are pages — no multiplier for scanned files.
Why image-only PDFs break normal translators — and the workflow that keeps your tables, columns, and stamps exactly where they were.
Read guideRandom symbols, blank pages, or 'PDF tools are temporarily offline' — the cause is almost always the same one thing, and it's fixable.
Read guideWe dropped a scanned PDF into Google Translate's document tab and it refused outright. Here's why — and the workflow that actually translates a scan.
Read guide