OCR analyses the pixel layout of a page and identifies characters. Recognition quality depends on scan clarity: 200–300 DPI is sufficient; text must be straight and high-contrast. Skewed or overexposed pages will be recognized less accurately.
After processing you get a PDF with 'dual' content: the page image remains, but a text layer sits beneath it. This lets you search (Ctrl+F), copy snippets, and allows screen readers to read the document aloud.
The tool supports Russian and English. For mixed-language documents, selecting the primary language is usually sufficient for accurate recognition.