search

tesseract-ocr / tessdata_best

Best (most accurate) trained LSTM models.
Apache License 2.0
1.26k stars 387 forks source link

readme

tessdata_best – Best (most accurate) trained models

This repository contains the best trained models for the Tesseract Open Source OCR Engine.

These models only work with the LSTM OCR engine of Tesseract 4.

See the Tesseract docs for additional information.

All data in the repository are licensed under the Apache-2.0 License, see file LICENSE.