Bounded inspection
Reports page count, native dimensions, DPI and gamma when exposed, bundled structure, hidden text, annotations, thumbnails, and foreground, background, and mask layers.
Recover the hidden OCR text already embedded in a DjVu document as UTF-8 plain text or bounded positional XML. This is extraction, not new OCR.
Private results · no network lookup · no quality reduction · files up to 1 GB
Reports page count, native dimensions, DPI and gamma when exposed, bundled structure, hidden text, annotations, thumbnails, and foreground, background, and mask layers.
Text accuracy and language come entirely from the uploaded source. Convertr does not improve, translate, mutate, or send extracted text upstream.
Selected pages are decoded by the pinned ddjvu build at their full native dimensions. Unsafe expansion fails explicitly instead of truncating or downscaling output.
Indirect DjVu documents with unresolved component references are rejected. The toolkit never follows component URLs, exposes component identifiers, accepts command scripts, or offers crop and quality switches.
Recover only the hidden OCR text already embedded in a DjVu document.
A missing text layer, invalid UTF-8, unsafe XML expansion, malformed document, or workspace exhaustion produces an explicit failure.
Upload a DjVu with embedded text and download the exact djvutxt UTF-8 output.