OCR image or scanned PDF
Make a scan searchable or extract English text from an image. Use OCR image or scanned PDF locally with practical instructions, output settings and format limits.
OCR image or scanned PDF
Recognize English text in a still image or scanned PDF and export plain text or a searchable PDF. Searchable output retains the page image with an aligned text layer, making it useful for finding words in scanned material. Review the recognized text, especially names, numbers, small print and unusual layouts.
- Make an English scanned PDF searchable
- Extract draft text from a photographed page
- Recognize printed English text from a still scan
How to use OCR image or scanned PDF
Choose a clear scan
Add one supported still image or an unprotected scanned PDF with no more than twenty pages. Upright text and clear contrast improve the recognition task.
Choose text or searchable PDF
Select TXT for recognized words or PDF to retain page images with an aligned text layer. This OCR workflow uses the local English recognition model.
Recognize the page content
Run Process document and wait for the pages to be rendered and recognized. Avoid treating completion as evidence that every word or reading order is correct.
Review against the scan
Download the result and compare meaningful passages with the original. In searchable PDF output, test search and text selection as well as the visible page image.
Understand the settings
- Output
- Choose output according to the result you need. Available choices: TXT; PDF. The initial selection is TXT.
A searchable English archive scan
A ten-page English instruction packet exists only as scanned pages. Choose PDF output and run OCR. Open the searchable copy, search for a known heading and verify a few selected passages against the visible image before using extracted text elsewhere.
Check your result
Choose a file, adjust the settings, then download the result. Local English OCR. Searchable PDF includes the original page image and an aligned text layer. Maximum 20 pages. Open the saved output in the app or workflow where you plan to use it. Check one representative result before repeating the same settings with more files. The original files stay unchanged.
What to know before you start
- Local English OCR. Searchable PDF includes the original page image and an aligned text layer. Maximum 20 pages.
- Upload one file with a combined size of at most 100 MB. The completed output must fit within 200 MB.
- Files are checked by the local engine; a matching filename alone is not proof of valid contents.
- Jobs run one at a time with a 180-second deadline. Temporary files expire after approximately 30 minutes; download them before clearing or restarting the app.
OCR image or scanned PDF: common questions
Which language does this OCR support?
This workflow uses local English OCR. Other languages or mixed scripts may produce poor recognition and should not be assumed supported. Choose a suitable language-specific OCR workflow when accurate recognition depends on a different writing system.
Will searchable PDF replace the scan image?
No. Searchable output includes the original page image representation together with an aligned recognized-text layer. The visible scan remains useful for comparison, while search and selection depend on the accuracy and placement of the recognized text.
How many PDF pages can be recognized?
The render-and-OCR workflow accepts up to twenty pages. For a longer document, prepare a relevant smaller PDF selection first and review recognition in manageable sections rather than expecting the full one-hundred-page PDF limit here.
Can I trust every recognized number?
No. OCR can confuse similar characters, punctuation, table columns and faint marks. Check meaningful quantities, dates, names and identifiers against the page image before relying on them, especially when copying the result into another document or table.
Does OCR image or scanned PDF upload files to another service?
Processing happens on the local server on this computer, using installed libraries, models and native tools. The browser-to-localhost connection is required, but no third-party conversion service is used. Initial dependency setup can require internet access; conversions use the installed local assets.
Can I continue after changing pages?
Pending selections stay available while navigating inside the app. Use My files to return to a tool draft or its results. Reloading or closing the tab discards pending browser selections; temporary job results expire after approximately 30 minutes. Download finished files before restarting the local app.