Importing a document

After “New case” you select the source document. Besides PDF, all common Office and image formats are supported (Word, Excel, images, TIFF and many more) — the document is first converted into a PDF rendition and then captured with text recognition (OCR):

Progress runs in the footer; Cancel (Esc) aborts the import. How long it takes depends on the number of pages, the OCR languages selected and the text source (see Settings → Text recognition). When it finishes, the status bar shows the page and word count, for example “Done — 2 pages, 450 words”.

Supported formats

pdf, docx, xlsx, pptx, odt, txt, csv, log, eml, msg, html, htm, png, jpg, jpeg, tif, tiff

Word, Excel, PowerPoint and text files can later be saved in the original format again.

Existing text or text recognition

Many PDFs already carry a text layer — searchable scans, for example. Whether the application takes over this existing text or reads the pages afresh with text recognition is governed by the Text source setting. In the default setting Automatic, existing text is taken over and supplemented by text recognition for pages without text and for images on the pages. The remaining variants are described under Settings → Text recognition. A changed setting takes effect from the next import onwards.

What happens in detail

During import, the document visibly passes through the phases Convert to PDF → Normalise pages → OCR (text recognition) → Extract text → Generate page previews → Analyse document.

Important

Text recognition is the basis of every word tool. Handwriting, stamps or heavily degraded scans that OCR does not capture cannot be redacted as a “word” — there, rectangle redaction helps, and the Show detections switch reveals where OCR found nothing.