AutoTag PDF (Paddle)

Start Automating Accessibility with Paddle AI – 100% Local, 100% free solution that transforms PDF remediation with enterprise-grade accuracy, zero cloud dependency, and complete data privacy.

PDFix now integrates with Paddle-based OCR and vision models (PaddleOCR & PaddleX), giving you a Dockerized solution to auto-tag PDFs and generate reusable layout templates. Together, PDFix SDK and Paddle turns complex, untagged PDFs into screen reader–friendly documents entirely on your own infrastructure.

Meet PDFix Desktop + Paddle: Auto-Tag PDF with Local AI

• 100% Free to Start: Open-source Docling model + PDFix Desktop integration
• Runs Localy: Process sensitive PDFs on your own machine via Docker
• PaddleOCR: Built on Paddle’s multilingual OCR and layout analysis toolkit
• Layout & Structure Detection: Leverages Paddle’s layout and table detection models
• Accessibility: Auto-tag PDFs for headings, paragraphs, lists, tables, figures, and formulas
• Templates: Automatically generate PDFix layout templates from Paddle’s layout analysis
• Batch Process-Ready: Trigger the action across files for high-volume PDF accessibility

How It Works: Choose Your Path

For PDFix Desktop Users

Perfect for accessibility professionals and document specialists who prefer a visual workflow while still using local AI models.

• Download PDFix Desktop
• 🐳 Install Docker Desktop First!
• Open PDFix Desktop and pull the container in the Action Manager (one-time set up) → AutoTag (Paddle)
• Upload a PDF to PDFix Desktop → Run action

For PDFix SDK Users

Perfect for developers, integrators, and enterprises who want to embed Paddle-based AI OCR and PDF remediation into automated workflows.

• Automated Pipeline Integration
  • Use the PDFix SDK to call the Docker action programmatically.
• Resources for SDK Integration
  • 📦 Docker Hub: https://hub.docker.com/r/pdfix/pdf-accessibility-paddle
  • 💻 GitHub Repository: https://github.com/pdfix/action-pdf-accessibility-paddle-docker
• SDK Benefits
  • Programmatic control over auto-tagging and template creation.
  • Integrate directly into existing content pipelines
  • Batch-process large volumes
  • Keep all content inside your network

🖥️ Local Processing -> PDFix + Paddle

• Complete Data Control – No PDFs leave your environment
• Unlimited Processing
• Works Offline
• Predictable Costs

Compared to cloud-only “auto-tag PDF” services, PDFix + Paddle gives you an on-prem, AI-powered PDF remediation stack that you fully own and control.

The Technology Behind It

• PaddleOCR
  • Multilingual OCR toolkit built on PaddlePaddle, designed to turn PDFs and images into structured data
• PaddleX
  • Low-code toolkit built on PaddlePaddle that streamlines model training, fine-tuning, and deployment with pre-trained models
• PDFix SDK
  • The Dockerized solution connects Paddle’s AI models with the PDFix remediation engine
• Template System
  • You can either use the action directly for automatic tagging or add an intermediate step by generating a layout template, which you can review and adjust if needed. Read more about Layout Templates.

Resources

• Getting Started Guide: https://pdfix.net/user-guide-external-actions/
• GitHub Repository:  https://github.com/pdfix/action-pdf-accessibility-paddle-docker
• Docker: https://hub.docker.com/r/pdfix/pdf-accessibility-paddle

Actions