OCR & file support
Scanned pages become searchable evidence
Pramvex reads the formats your work arrives in. Text in scanned PDFs is recognised on the device, so nothing is sent away for OCR.
Supported files
- PDF, both digital and scanned
- Word (.docx)
- PowerPoint (.pptx)
- Plain text (.txt) and Markdown (.md)
How OCR works in Pramvex
When a PDF page has no text layer, Pramvex runs optical character recognition on it locally, using the open PaddleOCR models (Apache-2.0) through an on-device runtime. The recognised text is indexed like any other text, and results that came from OCR are labelled OCR so you know to check the original page image.
Limits
- OCR accuracy depends on scan quality. Faint, skewed or handwritten pages may be read imperfectly.
- Password-protected or damaged files can’t be read. The Jobs view shows which files failed and why.
- In the proposed plans, OCR is part of Pramvex Pro.
Find the answer. Verify the evidence. Keep your files private.
Free to start. No account, no upload, no subscription.