Because the recognition happens on your machine, the scans that most need OCR — contracts, identity documents, records — are never transmitted anywhere, and the tool keeps working with Wi-Fi off once the app has loaded. Quality follows the scan: straight, clean pages recognize well; skewed or faded ones benefit from Clean scanned PDF first (run it before OCR, since cleanup rebuilds pages as images). Prefer the raw words? Choose plain-text output and get a .txt with a marker per page. Check important names and figures against the page image, which is unchanged.
How it works
- Open Bindery in your Mac browser and drop the scan into OCR
- Choose the language and searchable-PDF (or plain-text) output
- Download; test in Preview with Command-F
Questions people ask
Does it work offline on a Mac?
Yes. The engine and its language packs are part of the app and load from Bindery, not from a cloud service; once loaded, OCR runs with the network disconnected. That is also a simple way to see for yourself that the pages are not being uploaded.
Will Spotlight find the text afterwards?
Yes. The searchable output is a standard PDF with a text layer, which Spotlight indexes like any other PDF, and Preview’s Command-F finds words in it. The page images are pixel-identical to the scan, so nothing visibly changes. Plain-text output is also available if you only need the words.
Which languages are available?
Eleven: English, Spanish, French, German, Portuguese, Italian, Arabic, Simplified Chinese, Japanese, Russian and Hindi. Choose the document’s language before running; the engine relies on the language pack to resolve ambiguous shapes, and the wrong pack costs accuracy. Packs load from Bindery the first time each is used.