Scanned PDF Converter
Turn a scanned or image-only PDF into a searchable, selectable-text PDF with OCR — pick a language, optionally keep just the text layer to shrink the file. Runs entirely in your browser; your file is never uploaded.
Input
Higher resolution improves OCR accuracy on small text, at the cost of a larger output file.
Searchable PDF
Upload a scanned or image-only PDF and press Make searchable. Pages are rendered and OCR'd in your browser and stitched into a downloadable, selectable-text PDF. Your file is never uploaded.
Guides
Scanned PDF Converter turns a scanned or photographed PDF — where every page is really just a flat image — into a PDF with real, selectable text underneath. Search it, copy a paragraph out of it, or run it through a text extractor, all without retyping a thing. Everything runs in your browser with OCR; nothing is uploaded.
How to use it
- Drop or browse for a PDF file.
- Pick the document's language so the OCR engine reads it accurately.
- Optionally raise the render resolution if the scan has small or faint text — higher resolution reads more accurately but produces a larger file.
- Optionally limit the conversion to a page range like
1-3, 5. Leave it blank to process every page. - Turn on text-only output if you don't need to keep the original page images — it drops them and keeps just the invisible text layer, for a much smaller file.
- Press Make searchable, then download the result.
How it works
The conversion happens entirely on your device in three steps: pdf.js (the engine behind Firefox's PDF viewer) renders each page to an image, Tesseract OCRs that image and produces a one-page PDF with the recognized text laid invisibly over it, and the per-page PDFs are stitched back into a single document. Your file never leaves your browser.
Will it keep my original page images?
By default, yes — the output looks identical to your original scan, just with a hidden, selectable text layer behind each page's image. Turn on text-only output if you'd rather have a much smaller file with just the recognized text (the pages will render as plain text instead of the scanned image).
Why does OCR accuracy vary?
Recognition quality depends on the scan itself — a clean, high-contrast, non-skewed page reads far better than a blurry photo or a low-resolution fax. Raising the render resolution helps with small text; picking the correct language matters most for anything beyond basic English.
Is this the same as full text extraction?
Not quite — Image to Text Converter pulls the recognized text out as plain text you can copy directly. This tool instead embeds that recognized text back into the PDF itself, so the file stays a normal, shareable PDF that just happens to be searchable and selectable.