Does it perform OCR?
No. Image-only or scanned pages are identified as having no selectable text, but are not transcribed.
Inspect selectable PDF text and structure, report extraction confidence and export text, JSON or detected tables as CSV.
Embedded text, page statistics, metadata and reliably aligned tables only.
Choose one text-based PDF to inspect locally.
The PDF stays in this browser. Maximum 5 MiB and 25 pages. Scanned pages require a separate approved OCR workflow.
Extracts embedded PDF text and basic metadata, detects aligned text tables and labels page-level confidence.
Input: A digitally generated report containing selectable text
Output: Page text, metadata, table candidates and structured exports
No. Image-only or scanned pages are identified as having no selectable text, but are not transcribed.
It is a transparent heuristic based on embedded selectable-text coverage, not a claim that the extracted content is factually correct.
No. Only consistently aligned text rows are reported. Complex, merged or graphical tables may be missed.