English Čeština Deutsch Magyar Français Español Português (Brasil)

OCR

2 min read

On this page

How to use it

  1. Review Suggested languages and use Choose languages to select the languages in your document.

  2. Use Pages to recognize to choose the pages to process.

  3. Enable Re-recognize text only when selected pages that already contain text should also be processed.

  4. Run recognition, then download and test searching or copying a known phrase.

Choose the language of the page

OCR reads visible letters and adds recognizable text. The selected language affects recognition.

  1. Read the suggested languages.

  2. Choose Languages when the page uses a different language.

  3. Select the language used in the document. Read any language-pack download information before running OCR.

Choose the language of the page

Recognize only the pages you need

A small page selection is useful for checking recognition before a larger run.

  1. Enter 1 to test the first page.

  2. Use 1, 3 for separate pages or 1-3 for an interval.

  3. Leave the field empty when all pages should be processed.

Recognize only the pages you need

Replace an existing recognition layer

Re-recognize text also processes selected pages that already contain text.

  1. Enable Re-recognize text only when you want those pages processed again.

  2. Check the selected language and page range.

  3. After downloading, search for a known word and copy a sentence. Correct recognition matters more than a successful status.

Replace an existing recognition layer

Download and check the result

The result view shows the created output. Downloading a file and checking its contents are separate steps.

  1. Read the output name, count and size in the result view. If processing reported a problem, return to the settings and correct it before continuing.

  2. Choose Download and keep the new file separate from your original. If several outputs are packaged together, open the archive to find them.

  3. Open the downloaded result in a suitable viewer. Check the pages and content affected by this task. Keep your original until you are satisfied.

Download and check the result

Output and limits

OCR adds a searchable text layer to page images. It does not turn the layout into a word-processing document. The complete PDFs are saved, including untargeted pages. The tool is marked Beta and recognized text may contain errors. Check names, numbers and language-specific characters. Poor focus and complex backgrounds reduce reliability.

Image preview

Confirm