OCR backend for Regira Office, built on the Tesseract wrapper for Tesseract 5. OcrManager implements IOcrService and reads the text in an image, in the language its Options name (en by default) or the one a call passes.
<PackageReference Include="Regira.Office.OCR.Tesseract" Version="6.*" />
tessdata/eng.traineddata, the default Options.DataDirectory. For another language, add its .traineddata file from tessdata or the smaller tessdata_fast to that folder, or point DataDirectory at a folder of your own. Set the MSBuild property RegiraTesseractCopyTessData to false to leave the English data out.DataDirectory is relative to the working directory when the folder exists there, and to the output folder otherwise, so the default finds tessdata/ under dotnet run too. An empty one leaves the folder to Tesseract’s TESSDATA_PREFIX.IOcrService contract, the result model, and how the Tesseract and PaddleOCR backends compareApache License 2.0 — this package contains no license validation and no runtime limits. See LICENSE. A few companion packages are commercially licensed with a free tier; see the licensing overview.