Practical guide
A good file import starts before upload: choose readable source material, define the language direction, and decide how much vocabulary you can realistically review.
Guide steps
Text-based documents produce more reliable candidates than image-only scans. If text cannot be selected or copied from a PDF, extraction may be incomplete and a text or ebook version is usually a better source.
Set the language of the source and the language you want on the other side of each card. Keeping one direction per module makes later editing and review easier to understand.
A whole book can contain far more vocabulary than one review queue needs. Use the import flow to divide long material into smaller modules built around a chapter, topic, or reading session.
Imported vocabulary becomes part of your personal library. Check ambiguous translations, preserve useful examples, remove terms you already know, and then begin spaced review.
Questions
The current file import flow supports TXT, PDF, EPUB, and FB2. Results depend on the structure and text quality of the source.
An image-only scan may not expose enough text for reliable extraction. A text-based PDF, EPUB, FB2, or TXT version is a safer input when available.
You can start with a long source, but splitting it into smaller modules usually makes the resulting vocabulary easier to inspect and review.
Explore