OCR for scanned PDFs
Turn a scan into text
No database: AIMRAN doesn't store your files, text or results
What is OCR for scanned PDFs?
A scanned PDF is really a photo of each page: you can't search for words or copy the text. OCR reads those images and recognizes the letters to turn them into real text.
You get the recognized text to copy or download and, if you like, a searchable PDF: the same document, looking the same, but one you can now search with Ctrl+F and select text in.
How to use it
- Drag in the scanned PDF.
- Choose the document's language: Spanish, English or both.
- Turn on “Create searchable PDF” if you want the PDF with text.
- Click “Recognize text” and wait for all the pages to be processed.
- Copy or download the text and, if you asked for it, the searchable PDF.
Advantages
- Turns scans into searchable PDFs without changing how they look.
- Recognition in Spanish, English or both.
- Processes every page of the document in one go.
- Free and no sign-up.
Use cases
- Make a contract or deed you only have as a scan searchable with Ctrl+F.
- Copy the text of a court ruling or an official notice that reached you as a scanned printout.
- Get the text out of photocopied lecture notes to edit it in Word.
- Turn a scanned journal article into a searchable PDF to find quotes quickly.
Technical details
Recognition uses tesseract.js with the LSTM engine, automatic page segmentation and preserved spaces between words. Each page is rendered with pdf.js at about 300 DPI, with a maximum of 4200 px per side, which is the resolution recommended for Tesseract. For the searchable PDF, Tesseract generates an invisible text layer per page that is overlaid with pdf-lib on the original page, without touching the image and respecting its rotation. Language data is downloaded the first time (about 2 MB for Spanish, 3 MB for English and 5 MB for both). Accuracy depends on scan quality; handwriting is not reliably recognized.
Frequently asked questions
How do I convert a scanned PDF into editable text?
Drag in the PDF, choose the language and click “Recognize text”. Then copy the text or download it as .txt to edit it.
What is a searchable PDF?
It is a PDF that looks just like the original scan but carries the recognized text hidden underneath, so you can search for words and copy passages.
How do I know if my PDF needs OCR?
If you can't select the text with your mouse or searching finds nothing, the PDF is scanned and needs OCR.
How do I get better results?
Scan at a good resolution, with straight, well-lit pages, and choose the document's correct language.
Related tools
- PDF to Word · Turn a PDF into an editable .docx with its formatting and images
- Extract text (OCR) · Read text from an image in Spanish or English
- Extract text · Copy all the text from the PDF
- Search in PDF · Find words and which page they appear on
- Clean text · Spaces, line breaks, emojis and HTML
- Merge PDF · Combine several PDFs and reorder pages by dragging