How to OCR a scanned PDF
OCR PDF recognises the text in scanned or photographed pages and adds it as an invisible layer under the image. The document looks the same, but now you can search for words, select and copy.
It is essential for archiving scans you want to find later, copying details from a paper contract, or preparing a document before converting it to Word or Excel.
Step by step
- Upload one or more scanned PDFs.
- Under Document language, choose the actual language, or a combination such as Spanish and English or Catalan, Spanish and English.
- Choose Accurate or Balanced and, if needed, adjust the background, text layout and what to do with text that is already recognised.
- Run OCR, try searching for a word in the preview, and download.
Settings that improve recognition
Language matters: accents and letters such as ñ, ç or the Catalan l·l are only read reliably if you set it. Accurate uses more complete models and double-checks uncertain readings; Balanced is faster and works well with sharp documents. Both straighten rotated and skewed scans.
Under Document background, pick Even paper for scanner output or Shadows or uneven lighting for phone photos. Under Text layout, Automatic detects pages and columns, A single block of text suits letters, and Sparse text is for labels, receipts and notes.
PDFs that already contain text
By default, pages that already have text are kept and only scanned pages are recognised. If the PDF has a poor OCR layer from another tool, choose Redo the previous OCR layer: that invisible text is replaced while visible digital text is kept.
Up to 50 pages per PDF and 50 MB per job. In a batch, the same settings apply to every file.
Frequently asked questions
- Which languages does it recognise?
- Spanish, Catalan, English, French, German, Italian and Portuguese, plus combinations such as Spanish and English or Catalan, Spanish and English.
- Does the PDF look different afterwards?
- No. The recognised text sits in an invisible layer; the page image stays as it was.
- Can it read handwriting?
- It is built for printed text. Clear handwriting may be partly read, but it is not reliable.
- Can I convert the scan to Word afterwards?
- Yes. PDF to Word detects scans and offers to run OCR in the same step.
- Does it work with photos taken on a phone?
- Yes. Choose the Shadows or uneven lighting background to improve reading.