PixelOkayPDF STUDIO / TRANSLATE
Translate the text in your PDF while keeping page dimensions, images and approximate text placement. Review, adjust and export entirely from your browser.
Import a text PDF to translate manually, or place your own text regions over a scanned page. No model is needed.
Review each region, mark it approved, and edit its PDF coordinates. Manually add regions to image-only scanned pages.
Check page layout for overflow and review warnings before PDF export.
Search source or translated text across pages.
Page dimensions, image placement and most non-text graphics remain in position. The editor paints translated text over the original text regions on a rendered page. It is not a semantic PDF editor or perfect font/layout reconstruction tool.
This version requires an extractable text layer. Image-only scans need OCR. Password-protected documents, complex rotated text, signatures and forms require a different workflow and may not be suitable.
Supported desktop browsers can use their built-in Translator model. Optional free OPUS-MT models download on demand for English ↔ Spanish and English ↔ French. The app does not upload PDF files to a backend.
No. It retains the page image and approximate text placement, but translations can be longer and need manual adjustment. Review complex tables, multi-column documents and colors.
The built-in browser Translator API is not universally available on mobile. The optional local OPUS models only support four English/French/Spanish directions. Manual editing works without a translation model.
No. This version creates image-based PDF pages to preserve the appearance. Export a text transcript or JSON alongside it if you need the extracted and translated text.
Import an extractable-text PDF, set the language pair and choose an available translation engine or manual review. Correct translations and fit regions before exporting selected pages. Save a review backup if you need to reopen the same document later.
Browser translation depends on secure context, feature availability and model downloads. OPUS-MT supports English↔French and English↔Spanish only; manual review supplies no machine translation. PDF.js downloads at import. Exported visual PDF pages are raster images without searchable text, forms, tags or signatures. Scanned-only, password-protected and rotated PDFs are unsupported. Backups contain document text and restore only against matching PDF fingerprint/positions/languages. Layout, translation accuracy and font matching require review.
References: PDF.js · Transformers.js
Import a PDF with extractable text, select a supported source and target language, and use an available browser translation feature or the configured limited OPUS-MT path. Review the translation and page fit before creating the visual PDF export.
Authored example: Test a short English paragraph translated to French, then compare names, numbers and line breaks against the original. Treat the sample as a review exercise, not evidence of translation quality for another document or language pair.
Limits and processing: PDF.js loads from a configured CDN. Browser translation is feature/model-dependent; the optional OPUS path covers English–French and English–Spanish directions. Scanned-only, rotated and password-protected PDFs are unsupported. The exported visual PDF is raster and not searchable; it does not retain forms, tags or signatures. Review backups contain document text.
No. The visual PDF export is raster, not searchable, and does not retain forms, tags or signatures.
No. The supplied OPUS paths cover English and French or English and Spanish in both directions. Browser translation depends on browser support and available models.