OCR PDF: recognize the text in scans
Recognize the text on scanned pages and add it, invisible, on top of each page: the PDF looks the same, but now you can search and copy its text. Everything happens on your device.
Processed on your deviceNothing is uploaded to servers
Drag one or more PDFs here
PDF files
Files are opened on your device and never uploaded.
How it works
Choose the scanned PDF
Drag the file into the workspace or use the button. You can choose several at once.
Choose the languages
Tick the document’s languages (English, Portuguese or Spanish) and the quality. The first time, MovaDoc downloads the recognition data: about 3 MB per language.
Download the searchable PDF
The pages look the same, with the recognized text hidden on top: now you can search and copy it. You can also download just the text (.txt).
Frequently asked questions
Are my files sent to a server?
No. Recognition runs in your browser, with the Tesseract engine. The first time, MovaDoc downloads the engine and the language data from its own site (about 3 MB per language), when you start using the tool (you tap or click it, or reach it with the keyboard) and before processing any file; your PDF never leaves your device.
Is the recognized text accurate?
It depends on the scan. Sharp, straight printed text is recognized almost without mistakes; crooked, blurry or low-resolution scans give more errors. Handwriting isn’t recognized. A page scanned sideways has to be rotated first (in “Organize PDF”). Always check anything important.
What changes in the PDF?
What you see stays the same: the recognized text is added, invisible, on top of each page, where each word is. That way you can select, copy and search the text. In tables, the words become searchable, but copied text loses the table layout.
Does it work offline?
Yes, after the first time: the recognition files are stored in your browser (in a private window, only until you close it). The first time (and when you pick another language or quality), you need an internet connection to download them.
Which languages does it recognize? And “Fast” or “Accurate”?
English, Portuguese and Spanish, even mixed in the same document. “Fast” is lighter and enough for sharp printed text; “Accurate” is better with small print and poor scans, but slower. On a computer, each page takes a few seconds; on a phone, it can take much longer.