01
Upload your files
Securely upload a scanned PDF, photographed page, JPG, PNG or TIFF file.
Scanned document translation
Translate scanned documents. Keep the formatting.
Translate scanned PDFs, photographed pages and image-based documents. Bluente uses OCR to recover the text, translate it and rebuild the result against the original layout.
Best for publishing, legal and corporate document teams.
Trusted by teams at
Preserve the layout of scanned document translations
Scanned documents contain critical information trapped inside pixels. Bluente recovers the content without turning the result into a reformatting project.
The page stays recognisable
OCR identifies text, reading order, tables and visual regions before translation begins. The translated content is then positioned against the source geometry, preserving the relationship between headings, paragraphs, figures and forms.
- Complex tables remain tables
- Multi-column reading order is reconstructed
- Stamps, seals and letterheads remain in place
Scan01
→
Translated01
Useful immediately after download
The result comes back as a searchable PDF or editable DOCX rather than another flat image. Review it against the source, share it with a colleague or move it into the next filing and documentation workflow.
- Searchable, copyable text layer
- PDF or DOCX output
- Bilingual output available for review
Original scan
Translated + searchable✓ ready for review
What is included in instant scanned-document translation?
One workflow handles recognition, translation and document reconstruction.
Formatting preserved
Original page structure, styling, image placement and document hierarchy remain attached to the translated content.
120+ languages
Translate scans across major languages and scripts, including CJK and right-to-left documents.
Built for professional files
Designed for contracts, exhibits, reports, publishing material, forms and other documents that still need to work.
Enterprise security
Zero data retention, encrypted processing and automatic deletion within 24 hours.
Multi-format input
Process scanned PDFs, JPG, JPEG, PNG and TIFF files through the same document engine.
Batch-ready processing
Move large files and multi-document jobs through one consistent workflow instead of page-by-page copy and paste.
How it works
Our streamlined process delivers fast, accurate and format-preserving document translations.
02
Select languages and settings
Choose the target language and any output or formatting preferences.
03
Receive the translated document
OCR, translation and reconstruction run together while the original layout is preserved.
04
Review and download
Download a searchable document ready for comparison, sharing, filing or editing.
What this is actually used for
Exhibits and evidence that arrived as scans
Court exhibits, faxed correspondence and filed documents rarely arrive as clean digital files. They arrive as scans of scans, and the translation has to survive being read alongside the original.
- Side-by-side bilingual output for comparison against the source
- Page order and pagination preserved across multi-page faxes
- Seals and stamps left intact rather than translated
Exhibit_C_scan.pdf34 pages
Scan
→
Searchable
Invoices, forms and screenshots
Text baked into an image is invisible to every text-first tool. Screenshots of an application, a photographed invoice, a form someone filled in and scanned — all of them become searchable, translated documents.
- JPG, PNG and TIFF as well as scanned PDF
- Returned as PDF with a searchable text layer, or as DOCX
- Tables in forms rebuilt rather than flattened
invoice_photo.jpgOCR
What you get back
| Source | Returned as | Notes |
|---|---|---|
| Scanned PDF, no text layer | PDF with searchable text, or DOCX | Translation placed over the original layout |
| JPG or PNG screenshot | PDF or DOCX | Invoices, forms, application screenshots |
| TIFF, multi-page fax | PDF or DOCX | Page order preserved |
| Photographed page | PDF or DOCX | Deskewed and corrected before recognition |
| Mixed scan and text PDF | PDF or DOCX | Only the image pages go through OCR |
Worth knowing before you buy
Test your worst case
A 300 dpi scan of printed text and a dim phone photo are not the same job. Send the exhibit nobody could read.
Image translation costs 5x credits
Recognition is heavier than text translation, and pricing reflects that. Worth knowing before you plan a bulk run.
OCR covers the full language set
All 120+ languages, including right-to-left scripts such as Arabic and Hebrew. Recognition is not restricted to a shorter list than translation.
Frequently asked questions
Does it work on handwriting?
Partly, and it is the most error-prone case on this page. Output should always be read by a human before being acted on. See handwritten prescriptions for that use case specifically.
Can I search the translated document afterwards?
Yes. The returned PDF carries a real text layer, so it is searchable and copyable rather than being another image.
What resolution do I need?
300 dpi for printed text is comfortable. Lower works, but recognition errors rise and they compound into translation errors.
Does my glossary apply?
Yes. Terminology is enforced on recognised text the same way it is on native documents. See terminology control.
Is the scan used to train models?
No. Documents are never used as training data, are encrypted in transit and at rest, and auto-delete within 24 hours.
How large a file can I send?
100 MB per file, the same limit as a native document — OCR jobs are not capped lower.
Send the scan that failed somewhere else
The useful test is the exhibit nobody could read.