PDF Text Recognition

A scanned PDF is basically a photo. To the eye it looks like text, but the computer only sees pixels - you cannot select anything, copy anything or search anything. OCR (optical character recognition) flips that around: it recognises the letters in the image and turns them into real, searchable text. That is exactly what PDF Text Recognition does, page by page and with previews so you can see what was recognised.

The focus is German government and finance mail. Tax notices, letters from the authorities, contracts, certificates, invoices - all of it often arrives scanned and is exactly then most annoying, when you want to reuse the numbers and deadlines inside. Ready-made presets set language and resolution to fit the document type so you do not have to fiddle with settings.

Recognition runs in German, English or a mix of both. German texts with umlauts and the sharp ß in particular benefit from choosing the right language. For documents that mix both languages there is the combined mode, which uses both dictionaries at once.

Resolution decides accuracy and speed. 150 DPI is fast and enough for clean, clearly printed pages. 300 DPI is the standard for most documents. 600 DPI squeezes out the maximum from small print, contracts or slightly blurry scans, but takes a bit longer. Every page comes with a confidence score that shows you how sure the recognition was.

You can view the result page by page or as one continuous text, search it right in the browser and pull the recognised text out - ready to reuse, say to copy a figure into a calculator or move a deadline into the Fristenwächter. Word and character counts per page are always in view.

The actual character recognition needs computing power and therefore runs on the server, but only for the pure recognition step. Your documents are not stored permanently. No subscription for a simple scan, no detour through a foreign cloud service with murky privacy terms.