How to use the OCR PDF tool
- Open the scanned PDFDrop the PDF onto the box or click to choose it.
- Choose the languagePick the language of the text. Keep “Also recognise English” on for mixed documents.
- Make it searchableEach page is read word by word, and an invisible text layer is placed exactly over the scanned text.
- DownloadSave the searchable PDF, or download only the recognised text as a .txt file.
Features
- Searchable & copyableUse Ctrl+F to find words, and select and copy text, in any PDF reader.
- Looks the sameThe scanned pages are not changed. The text layer is invisible and sits exactly over each word.
- 100+ languagesAll major world languages: English, Spanish, French, German, Arabic, Chinese, Japanese, Russian, Hindi, Bengali, Tamil and many more.
- Smart skippingPages that already contain text are left alone, so mixed documents are processed quickly.
- Text exportDownload all recognised text as a plain .txt file.
- PrivateOCR runs on your device with Tesseract. Your document is never uploaded.
What does OCR do to a PDF?
When you scan a document or save photos as a PDF, each page is just a picture. You can read it, but your computer can’t: there is no text to search, select or copy, and screen readers can’t read it aloud. Optical character recognition (OCR) looks at the picture, finds the letters and words, and turns them into real text.
This tool keeps your pages exactly as they are and adds the recognised words as an invisible text layer, positioned precisely over the words in the image. This is the same “searchable PDF” format that professional scanners and Adobe Acrobat create. You see the original scan; your PDF reader sees the text.
How it works
Each page is rendered at high resolution in your browser and read by Tesseract, a widely used open-source OCR engine developed with support from Google. The first time you use a language, its language data (a few megabytes) is downloaded and cached by your browser. Everything else, including your document, stays on your device, which makes the tool safe for contracts, ID documents and medical records.
Getting the best results
- Choose the right language. OCR uses language-specific dictionaries and letter shapes. For Hindi documents with English words, choose Hindi and keep “Also recognise English” ticked.
- Use “High” accuracy for small print, and for documents scanned at low resolution.
- Straight, clear scans work best. Rotate sideways pages first with Rotate PDF. Photos taken at an angle, with shadows, or with handwriting give poorer results.
- OCR is very accurate on clean printed text, but not perfect. Check important numbers, such as amounts, dates and account numbers, against the original.
What you can do with a searchable PDF
- Find any word in long scanned contracts, books or old records with Ctrl+F.
- Copy quotes and figures instead of retyping them.
- Let your computer, Google Drive or a document system index the content.
- Make documents accessible to people using screen readers.
- Convert the text further with PDF to Word, or extract it with PDF to Text.
Only need text from a few photos or screenshots? Use Image to Text instead.