Scanned PDFs need OCR before translation. Here are the 15 questions users ask most often — covering accuracy, layout, handwriting, supported languages, mobile photos, multi-page scans and safe handling of IDs and contracts.
Yes. TranslaPro automatically runs OCR (Optical Character Recognition) on scanned PDFs before translating. You don't need to convert the file first or run any separate tool — upload the scan, choose a target language, and the text is extracted, cleaned and translated in one step.
OCR analyses each page image, identifies character shapes, reconstructs words and paragraphs, and produces a digital text layer. TranslaPro uses a multi-tiered OCR pipeline that picks the best engine per language (Latin, Cyrillic, Arabic, CJK) so accuracy stays high even on low-resolution scans.
300 DPI or higher gives the best results. 200 DPI still works for clean printed text. Below that — or if the scan is skewed, dark or photographed by phone in poor light — accuracy drops. TranslaPro applies deskew, denoise and contrast correction automatically before OCR runs.
Printed text in scanned PDFs translates reliably. Handwritten notes are much harder: cursive, mixed scripts and unusual letterforms often fail OCR. For mixed documents (printed form + handwritten fields) the printed parts translate cleanly and the handwritten parts may need manual review.
TranslaPro re-paginates the translated content into a clean A4 PDF rather than overlaying text onto the scan. This is intentional: overlay-style tools usually break alignment, hide source text, or produce illegible output. A clean re-paginated document is far easier to read, sign or share.
OCR + translation is supported in 50+ languages, including all major European languages, Arabic, Hebrew, Chinese (Simplified & Traditional), Japanese, Korean, Hindi, Bengali, Tamil, Thai and Vietnamese. Right-to-left and CJK scripts are handled with dedicated OCR models for higher accuracy.
Yes. Scanned IDs, contracts, medical reports and bank statements are common use cases. Files are encrypted in transit, processed in isolated workers, never stored long-term, and auto-deleted within 30 minutes. We never train AI on your documents and never share them with third parties — fully GDPR compliant. See our security details and privacy policy.
Yes — TranslaPro is widely used for visa, USCIS, immigration and academic submissions. The output is a clean PDF you can attach to the application. For documents that require a sworn or certified translation, use the translated PDF as a working copy and have it certified by an authorized translator in your jurisdiction.
Almost always one of three reasons: (1) the scan is too low resolution, (2) the wrong source language was auto-detected, or (3) the PDF has a fake text layer (invisible OCR errors from a previous tool). TranslaPro re-runs OCR from scratch when it detects a corrupted text layer — so re-uploading the original scan usually fixes it.
Scanned PDFs add an OCR step, so they take longer than text-based PDFs — typically 1–3 seconds per page on top of translation time. Larger scans are processed in parallel chunks, so a 50-page scan still finishes in well under a couple of minutes in most cases. Pricing is per page regardless of OCR.
Yes. Convert the photo to a PDF first (iOS Files and Google Drive both do this in one tap), or combine multiple photos into a single PDF, then upload it to TranslaPro. Phone photos are usually lower contrast than flatbed scans, so make sure the page is well-lit, in focus, and shot straight from above for best OCR accuracy.
Yes. TranslaPro processes multi-page scans (up to ~200–300 pages, depending on density) in a single job and merges the results into one downloadable PDF. Each page is OCR'd independently and translated in parallel, so a 50-page scan typically completes in 1–2 minutes rather than 50× the per-page time.
Yes. After OCR extracts the text, TranslaPro runs language detection on the result to pick the source language — useful when you don't know what language the scan is in (common with foreign IDs, certificates and receipts). You can also override the detected language manually if the scan mixes two languages.
Pricing is the same as for normal PDFs — pay-per-document for one-off translations, or a monthly subscription for higher volumes. There is no extra OCR surcharge: the OCR step is included automatically when we detect a scan. See full pricing tiers.
OCR translation is fast (seconds to minutes), low cost, and great for understanding the content of a scanned document. A human translator is slower and more expensive but is required when you need a sworn, certified or notarised translation — for example, for court submissions or official immigration files. Many users start with TranslaPro to get a working draft, then send it to a certifier for sign-off.