JPG to Text (OCR)

Recognition runs on Tesseract compiled to WebAssembly, executing in a worker thread inside your browser. The first run downloads the trained data for the language you picked and caches it, so the initial recognition is noticeably slower than every one after it. No image is uploaded — the engine comes to the file rather than the file going to a server.

JPEG deserves one specific warning. It is a lossy format that compresses by discarding high-frequency detail, and the sharp edge between a black letter and a white page is exactly that kind of detail. A heavily compressed JPEG shows soft haloes and blocky fringes around characters, and those artefacts are what the recogniser has to work with.

If you control how the image is produced, capture text as PNG. If you only have the JPEG, a tighter crop around the text region usually helps more than any setting here.

The language setting matters more than people expect, because it is not a hint. Tesseract loads a different trained model per language, and each one encodes that language's character set and letter-sequence probabilities. Running Portuguese text through the English model does not degrade gracefully; accented characters have no good match and the engine's guesses about likely letter sequences are actively wrong.