Breadcrumbs

Module Autoscan

Autoscan is the AI-powered feature that extracts content from a PDF source document and populates your module's fields automatically. This saves manual typing and manual marking when your source content is already in a PDF.

When Autoscan is available

Autoscan is available whenever you're in the authoring canvas in two-column PDF mode — that is, when you created the module by choosing Upload a PDF. It runs alongside the manual text-marking option, so you can use whichever fits the section you're working on.

The two Autoscan modes

Autoscan can be run in two modes:

  • Autoscan a single page — extracts content from just the current page of the PDF. Useful when your source content is on one page and you don't want to process the entire document.

  • Autoscan the complete document — extracts content from the entire PDF at once. Best when your source document is short or when the module's content is spread across multiple pages.

What Autoscan does

Autoscan uses AI-powered OCR and content extraction to:

  1. Read text and structure from your PDF.

  2. Populate the existing fields of your module based on the content module type schema.

  3. Create new fields when it detects content that doesn't fit the existing schema — so you can capture information the type didn't anticipate.

Once extraction completes, you review and edit the fields in the authoring canvas. The PDF stays visible on the right so you can compare against the source at any time.

[Screenshot to add: Autoscan result — pre-populated fields alongside the PDF viewer]

Supported languages

Autoscan supports source documents in Latin character sets. Officially supported languages:

  • English

  • German

  • French

  • Spanish

  • Italian

  • Portuguese

Danish has also been verified to work.

Not supported

Autoscan does not extract non-Latin character sets, including Chinese, Japanese, Korean, Arabic, Hebrew, and other non-Latin scripts.

For modules sourced from non-Latin language content, use manual text input or the manual marking flow in the authoring canvas instead of Autoscan. All other module functionality is unaffected.