Un collègue m'a partagé le projet Marker (https://github.com/VikParuchuri/marker) :

Marker converts documents to markdown, JSON, and HTML quickly and accurately.

• Converts PDF, image, PPTX, DOCX, XLSX, HTML, EPUB files in all languages
• Formats tables, forms, equations, inline math, links, references, and code blocks
• Extracts and saves images
• Removes headers/footers/other artifacts
• Extensible with your own formatting and logic
• Optionally boost accuracy with LLMs
• Works on GPU, CPU, or MPS

source

Voici comment fonctionne Marker :

Marker is a pipeline of deep learning models:

• Extract text, OCR if necessary (heuristics, surya)
• Detect page layout and find reading order (surya)
• Clean and format each block (heuristics, texify, surya)
• Optionally use an LLM to improve quality
• Combine blocks and postprocess complete text

source