Why this page is focused on PDF.
PDF is the format where DocTranslate AI has the clearest product differentiation.
- Built for PDF structure.
The PDF workflow is layout-aware. It maps extracted text back into the document with position, spacing, and column structure in mind.
- Certification is PDF-only.
Only translated PDFs receive embedded integrity metadata and a verification page. Other formats stop at translation.
- Verification happens later on /verify.
Certified PDFs can be checked locally in the browser by recomputing the text hash and comparing it to the stored metadata.
What this PDF workflow still does not promise.
The PDF path is the strongest one in the product, but it still has practical limits.
- The rebuilt layout can still shift.
Tables, forms, uncommon fonts, and rotated elements may still move after rebuilding. The result is layout-aware, not pixel-perfect.
- Certification is not legal validation.
The PDF certificate proves file integrity only. It does not create notarization, sworn translation, or legal enforceability.
- Extracted text still leaves the browser for translation.
The original file stays local first, but extracted text is still sent to external services over HTTPS for anti-bot checks and translation.
PDF translator FAQ.
Common questions about the PDF-specific workflow.
Can DocTranslate AI translate a PDF without destroying the layout?
The PDF workflow is designed to preserve structure more carefully than plain text extraction. It works best on readable PDFs with consistent text layers, but some layout shift can still happen.
Does PDF translation include certification?
Yes. In this product, certification is available for translated PDFs only. The translated PDF receives embedded integrity metadata and an appended certificate page.
How do I verify a certified PDF later?
Open the /verify page and upload the certified PDF. The browser recomputes the text hash locally and checks whether it matches the metadata stored in the file.
What are the current PDF limits?
The public workflow currently limits PDF files to 10 MB and 30 pages. Very large or complex files may still hit processing or timeout limits.
Read how it works.Then translate.
The workflow page explains how parsing, chunking, translation, rebuilding, and PDF verification fit together.