Document OCR

Run OCR on scanned PDFs and images (JPEG, PNG, TIFF, BMP) up to 30 MB. Pick a Tesseract language, output plain text (.txt) or Word (.docx), and copy or download results after server-side processing.

ocrpdftexttesseractscanned pdf
OCR requires Tesseract to be installed on the server. Contact your server admin if this tool returns an error.

Upload a scanned PDF or image

Output format

What is Document OCR?

Document OCR turns image-only pages—scanned PDFs, photos of documents, or standalone images—into editable text. The tool sends your file through Dynamic Duniya’s file-tools API with a Tesseract-style language code, then shows extracted text (for .txt) or a download link for a Word document (.docx). It complements PDF-to-Excel and other converters: once a scan has a text layer, downstream tools can work with selectable content.

Supported inputs and limits

The uploader accepts PDF plus common raster formats: JPEG, PNG, TIFF, and BMP. The UI caps uploads at 30 MB. OCR is CPU-heavy; the progress label notes that large files may take on the order of 10–30 seconds. Success depends on Tesseract and related components being installed and healthy on the server—a blue info banner reminds operators to verify that setup if errors appear.

Languages and output

Pick a recognition language from the dropdown (for example English eng, Hindi hin, French fra, Spanish spa, German deu, Arabic ara, Chinese Simplified chi_sim, Japanese jpn, Portuguese por, Russian rus). Choose Plain Text (.txt) to read and copy text in the page, optionally with a download named like extracted-text.txt, or Word Document (.docx) for ocr-result.docx when the pipeline returns a file URL. Word counts may come from server meta or a local estimate.

Privacy

Unlike purely in-browser OCR, your document leaves the device for server processing. Do not upload regulated health, legal, or classified material unless policy allows. Download outputs promptly; temporary URLs can expire depending on configuration.

Frequently Asked Questions

Will OCR be perfect on every scan?

No. Blur, skew, low contrast, handwriting, watermarks, and complex layouts reduce accuracy. Always proofread critical numbers, names, and legal text.

My PDF already has selectable text—do I need OCR?

Usually not. If you can highlight words in a desktop reader, the PDF likely already has a text layer; OCR may duplicate or garble content. Use OCR when text cannot be selected or search fails.

Is OCR done in the browser?

No. The file is uploaded to Dynamic Duniya servers for Tesseract-based recognition, similar to other file tools on the site.

What is the maximum file size?

The dropzone is configured for up to 30 MB per file.

Should I choose TXT or DOCX?

TXT is ideal for quick copy-paste, scripts, or plain storage. DOCX is better when you want to open results in Word or share a formatted document; layout is still OCR-derived, not a faithful replica of the original design.

Why does the tool say Tesseract must be on the server?

Recognition runs server-side. If Tesseract or dependencies are missing or misconfigured, the API may return errors until an administrator installs or fixes them.

Is Document OCR free?

Yes for typical personal and work use; respect fair-use limits and your organization’s data rules.

Tips

Quick guidance for using our tools safely and effectively.

Privacy

Files are processed on the server for conversion only and are not used for training or shared with third parties.

Best results

Use the formats suggested in each tool. Large media files may take longer — keep the tab open until processing finishes.

Need something else?

Browse related tools below or explore other categories from the main Dev Tools hub.

Related tools

More utilities in the same category.

PDF to JPG / PNG

New

Export PDF pages as JPEG or PNG images in your browser (Mozilla pdf.js + canvas). Pick render scale, optional page list, and JPEG quality. Multi-page exports download as a ZIP. Max ~35 MB file size and up to 1000 pages (browser memory is the practical limit — use page ranges for huge jobs) — nothing is uploaded.

pdfimage+1

JPG / PNG to PDF

New

Stack JPEG or PNG files into one PDF with pdf-lib: reorder the list, choose A4 or Letter, set margins, and fit each image to the page — all client-side, no server upload. Up to 12 images per run.

pdfimage+1

PDF to DOCX

New

Upload a PDF (up to 50 MB) and convert to a downloadable .docx via Dynamic Duniya’s file API — best for text-heavy PDFs; multi-column layouts and images may not transfer cleanly. Progress while uploading; open the result in Word or Google Docs and review formatting.

pdfdocx+1

DOCX to PDF

New

Upload Word .docx or legacy .doc (up to 50 MB) and download a PDF via the server file API. Output fidelity depends on LibreOffice on the server — review complex layouts. Progress while uploading; suggested download name converted.pdf.

pdfdocx+1

Merge / Split PDF

New

Merge: first-page previews in a horizontal strip — drag cards to set merge order. Split: optional “Load page previews” then drag thumbnails (four per row, scrollable) to reorder pages before splitting by ranges or every N pages; ranges use preview slot numbers when previews are loaded. Zip when multiple parts. Client-side pdf.js + pdf-lib — max ~35 MB per file, 150 pages, 12 PDFs for merge.

pdfmerge+1

PDF Watermarking

New

Stamp text (font size, opacity, rotation, corner/center placement) or a JPEG/PNG image across every page using pdf-lib locally. Encrypted PDFs are not supported. Download a watermarked copy — your file never leaves the browser.

pdfwatermark