OCR PDF online: make scanned text searchable
Recognize printed words on scanned PDF pages and add a searchable text layer while retaining the original page images. OCR helps search and copy text that a normal PDF text extractor cannot read. Recognition quality depends on language, resolution, focus, contrast and the original scan.
100 MB per file · up to 500 pages per job · one job at a time. Files are temporary, expire within 24 hours and are cleaned up automatically. Download results promptly; service restarts can remove them earlier.
Your document is ready.
What OCR PDF does
Recognize printed words on scanned PDF pages and add a searchable text layer while retaining the original page images. OCR helps search and copy text that a normal PDF text extractor cannot read. Recognition quality depends on language, resolution, focus, contrast and the original scan.
How to use this tool
- Upload a scanned PDF and select pages and document languages.
- Choose Fast for clear scans or a 300 DPI preset for harder text.
- Follow actual completed pages and the estimate; download and test a phrase.
Common uses
- Make a scanned report searchable by a printed reference number after verifying the recognized digits.
- Add a text layer to an archive of typed paperwork while retaining the original page images.
- Copy a paragraph from an image-only PDF for authorized reuse, then compare the transcription with the scan.
- Process a representative excerpt first to select languages and quality before submitting a large scan.
Tips for better results
- Choose the languages actually printed in the document, up to four supported choices. Extra unrelated languages are not a substitute for a clearer scan.
- Fast uses 200 DPI for clear scans. Standard and High quality use 300 DPI for harder text. Compare a short representative page selection before processing the whole report.
- Test a phrase with search and copy an important number into a text editor. OCR can misread punctuation, dates and similar-looking characters even when the page image looks correct.
- The progress count reports completed worker pages. Its estimate covers page processing; upload, output saving and validation add time and may not advance the count.
Formats and limits
- Input
- Output
- Current file limit
- 100 MB per file, up to 500 pages per job, one job at a time. Shared usage quotas apply.
Supported languages come from the current backend. OCR is not a promise of perfect transcription, handwriting recognition, translation or a fixed completion time.
Questions about OCR PDF
Does OCR change the visible scan?
The OCR pipeline retains the original page images and adds recognized text. Check the result if the document is particularly important.
What does the time estimate include?
The displayed estimate covers page processing. Uploading, saving and output validation can take additional time.
Why can I see words but not select them?
An image-only scan displays photographs of text. OCR recognizes printed words and adds a searchable layer; a normal text extractor cannot read that image by itself.
Does Fast preserve the original page image?
The OCR pipeline retains original page images in every preset. Fast changes recognition resolution, so compare recognized small text rather than assuming equal transcription accuracy.
Is there a fixed time for a 200-page PDF?
No. Scan quality, selected languages, image size and shared CPU capacity affect time. The displayed estimate uses completed pages and excludes some upload and finishing work.
Does OCR translate or guarantee handwriting recognition?
No. Choose supported recognition languages for printed text. Translation, perfect transcription and reliable handwriting recognition are not promised features.
Can I recognize Hindi text in a PDF?
Hindi OCR is not currently supported. Use the available language list for printed text; English does not reliably recognize Devanagari. Image conversion can still package photos of Hindi pages.
Your files and privacy
Documents are uploaded over HTTPS to the hosted service. Files are temporary and protected by your browser session. Download results promptly and keep a local copy before temporary files expire.
Read document handling details · Review security and processing limits
Related PDF tools
Practical guide
Identify image-only scans, select an OCR preset and language, follow page progress and check recognized text against the original.
How to make a scanned PDF searchable with OCR