Tesseract language models (spa, eng) — "best" LSTM models, integer-quantised
(4.0.0_best_int), as packaged by @tesseract.js-data/spa@1.0.0 and
@tesseract.js-data/eng@1.0.0 (npm; packaging licence MIT, Copyright Balearica /
jeromewu, https://github.com/naptha/tessdata).

The model data comes from https://github.com/tesseract-ocr/tessdata_best,
Copyright the Tesseract OCR authors, licensed under the Apache License,
Version 2.0 (the full text ships in ../tesseract-7.0.0/LICENSE_TESSERACT_JS_CORE):
http://www.apache.org/licenses/LICENSE-2.0

Files are byte-identical to the npm packages (sha256 pinned in
web-app/tests/ui/ocr-engine.test.js).