This package is designed to take images, pdfs, or raw text, extracts the text, then corrects it using LLMs. There are additional options to collate the pages of text with various properties (e.g., eliminating line breaks, headers and footers, etc.), and translation.
For usage, please see the walkthrough.ipynb notebook under notebooks/. A streamlit interface is forthcoming under the streamlit/ directory. And, more documentation is also in the works.