Important Privacy Notice

Please do not upload confidential or personal data.

All uploaded PDF files are temporarily stored for processing and automatically deleted after 30 days. For maximum data security, we offer special solutions.

Secure Local Mode available

For sensitive documents, we offer a protected processing mode that can be provided internally and locally:

  • On-Premise Installation in your organization
  • No cloud processing – all data stays with you
  • GDPR-compliant solution for public authorities and companies
  • Full control over your sensitive documents
Request Secure Mode

Email Support

Professional support via email – response within 48 hours on business days for all your questions about PDF Accessibility Pro.

Response time: 48 hours on business days
Send email: info@barrierefrai.com

Processing Guarantee

We want you to be 100% satisfied with the results. If AI processing does not meet your expectations (e.g. wrong text order, formatting errors, or unsuitable image descriptions), we will review the result manually.

Guaranteed: We improve the tool and you will receive your credits back.
Request manual review

Current Pricing Structure

🆓 Free Trial
Free
25 pages free
Single-page processing only
Max. 5 pages/document
HTML & PDF/UA export
📄 Starter
99€/month
250 pages monthly
Single-page + batch processing
Live progress display
Extended PDF/UA
⭐ Pro
199€/month
600 pages monthly
Advanced text conversion
Optimized PDF/UA structure
PDF/UA Hybrid – layout and text version combined
Priority support
🏢 Enterprise
499€/month
1500 pages monthly
Batch processing
PDF/UA Hybrid – layout and text version combined
Secure Local Mode
Premium support

Quickstart Guide

Create accessible PDFs in 3 easy steps

1
Upload PDF

Drag your PDF file into the upload area or click to select.

2
Select processing mode

Single-page (up to 5 pages) or batch processing (from 5 pages) with live progress.

3
Download

Download the accessible HTML or create a PDF/UA document.

Frequently Asked Questions (FAQ)

💎 Current Pricing Structure

What does the new pricing model offer?

Our new, transparent pricing structure (from 2024):

  • Free Trial: 25 pages once free to test (single-page processing only, max. 10 pages/document)
  • Starter (99€/month): 250 pages, single-page and batch processing
  • Pro (199€/month): 600 pages, extended features, accessible text version and PDF/UA Hybrid
  • Enterprise (499€/month): 1500 pages, batch processing, PDF/UA Hybrid, Secure Local Mode

New: Every user receives 25 free pages to test!

Flexible: You can cancel monthly until renewal, and unused credits are transferable.

How does the page counting work?

Consumption-based billing:

  • Each processed PDF page counts as 1 page from the monthly quota
  • Standard mode: Text extraction only = 1 page per PDF page
  • Compact/Detailed: AI analysis = 1 page per PDF page
  • Vision mode: Image analysis = 1 page per PDF page (Pro/Enterprise only)

Example: A 10-page PDF uses 10 pages from your monthly quota, regardless of the selected analysis mode.

🔄 PDF Processing

Which processing modes are available?
Single-Page Processing

For: PDFs up to 5 pages

Speed: Very fast

Method: Immediate conversion

Available in: Free Trial, Starter, Pro, Enterprise

Feature: Direct result without waiting time

Batch Processing

For: PDFs from 5 pages

Speed: Varies depending on page count

Method: Page-by-page processing

Available in: Starter, Pro, Enterprise

Feature: Live progress per page

Automatic selection: The system automatically chooses the appropriate mode based on the number of pages in your PDF.
How does batch processing work?

Batch processing shows the progress of each individual page:

📊 Live progress display
  • Page overview: All pages as thumbnails
  • Status per page: “Ready” → “Processing” → “Done”
  • Color coding: Grey → Orange → Green
  • Real-time updates: Immediate status changes
  • Overall progress: Percentage for the whole document
⚙️ Processing steps
  1. PDF is split into single pages
  2. Each page is converted individually
  3. Text extraction and HTML structuring
  4. Reassembly into a complete document
  5. Download of the finished accessible HTML
Transparent: You can see exactly which page is being processed and follow progress in real time.

📄 PDF/UA Export

What is PDF/UA and why is it important?

PDF/UA (PDF Universal Accessibility) is the international standard (ISO 14289-1) for accessible PDF documents.

Important note on PDF/UA conversion

The PDF/UA creation produces a text-based, accessible version of your document specifically for users with visual impairments and screen reader users.

The original PDF is NOT converted into an accessible version – instead, a new, structured text version with an HTML base is created, optimized for assistive technologies.

✅ What PDF/UA provides:
  • Screen reader compatible: Optimized for NVDA, JAWS, VoiceOver
  • Navigable structure: Heading navigation
  • Logical reading order: Correct text sequence
  • Structured content: Lists, tables, paragraphs
  • Text-based access: No visual barriers
🏛️ Legally required for:
  • Public administration (BITV 2.0)
  • Educational institutions
  • Large companies (from 2025)
  • EU institutions (EN 301 549)
  • Accessible communication
Target group: PDF/UA is specifically designed for people who rely on screen readers and other assistive technologies.

🪄 PDF/UA Hybrid

What is PDF/UA Hybrid and how does it differ from the accessible text version?

The key difference:

Standard PDF/UA Export
  • Creates a new accessible text version
  • Layout may differ from the original
  • Two separate documents
  • Optimised for screen reader readability
PDF/UA Hybrid
  • One single file – no second document
  • Visible: exactly the original layout
  • Invisibly embedded: complete PDF/UA layer
  • Passes all PAC-3 & screen reader tests
Principle: Add, never change – The original text and layout remain completely untouched. The accessibility layer is only added on top – nothing is ever overwritten.
Why is PDF/UA Hybrid PDF/UA-compliant and how is it tested?

The technical foundation:

Our singularly developed server technology embeds a complete semantic structure layer directly into the PDF – in accordance with ISO 14289-1 (PDF/UA standard). This layer is invisible to the viewer but fully readable by screen readers and validation tools.

📋 What is embedded:
  • Tagged PDF structure (ISO 14289-1)
  • Role mapping: H1–H6, P, Table, List, Figure…
  • Reading order (logical text sequence)
  • Alt-texts for all images and graphics
  • Document language and metadata
🔬 Why PAC-3 passes this test:
  • PAC-3 checks only the structure tags, not the visual layout
  • Tagged PDF and visual rendering are completely separate layers in PDF files
  • The original layout does not interfere with the validation
  • Screen readers only read the structured layer
Key insight: PAC-3 validates the structure tags, not the appearance – that is exactly what the Hybrid approach leverages. The visible layout layer and the invisible PDF/UA layer coexist completely independently.
Why does PAC show Quality notes even though PDF/UA and WCAG passed?

PDF/UA Hybrid combines the original visual layout as a protected layout layer with the accessible text version. This preserves design, line breaks and corporate layout while screen readers use the accessible structure.

Some PAC Quality and AI notes additionally use heuristics to compare the visible layout geometry with the structure tree. In PDF/UA Hybrid files, PAC can therefore show notes even though the normative PDF/UA and WCAG checks have passed.

A typical example is the Quality note “Link in text does not have a Link element”. This can occur when PAC detects link-like text in the visual background layer. The link semantics that matter for screen readers, keyboard navigation and PDF/UA are stored in the accessible structure layer of the document.

This means: if the PDF/UA, WCAG and ISO checks are green and the links work in the accessible structure, this Quality note must be assessed in context. It can result from the PDF/UA Hybrid architecture.

Important: The decisive conformance checks are PDF/UA, WCAG and ISO. Quality notes are reviewed separately and manually assessed when they indicate a real usability issue.
Why does PAC show AI warnings for PDF/UA Hybrid?

In addition to the normative PDF/UA, WCAG and ISO checks, PAC also includes AI-assisted quality notes. These AI notes do not only validate the formal PDF/UA structure. They heuristically assess whether the visible layout, semantic structure and expected reading logic appear to match.

With PDF/UA Hybrid, PAC may therefore show AI warnings even though PDF/UA, WCAG and ISO have passed. This is caused by the specific architecture: the visible original layout remains in a separate background layer, while screen readers use the separate accessible text version.

Important: AI warnings are quality assurance notes, not automatic accessibility failures. The decisive technical conformance checks are PDF/UA, WCAG and ISO. AI notes should still be reviewed, especially for legally relevant documents, complex tables, forms or unusual reading order.
Is PDF/UA Hybrid really 100% accessible?

An honest assessment:

Our AI automatically handles ~90% of the accessibility work – structure tags, reading order, alt-texts for images, table descriptions. The result passes PDF/UA certification (PAC-3).

✅ Handled automatically:
  • Heading hierarchy (H1–H6)
  • Logical reading order
  • Alt-texts (AI-generated via GPT-4o Vision)
  • Table & list structure
  • ISO 14289-1 / WCAG 2.1 AA compliance
🔍 Recommended manual review:
  • Very complex diagrams and infographics
  • Specialist or scientific illustrations
  • Multi-column special layouts
  • Legally relevant or sensitive content
Key takeaway: AI handles ~90% of the work for you. The remaining 10% – especially for very complex content – determines perfect quality and should be spot-checked.
What are the concrete advantages of PDF/UA Hybrid over previous solutions?

5 decisive advantages at a glance:

  • No layout loss – print approvals and corporate design remain unchanged
  • One single file – no version confusion, no “_accessible” suffix, no complexity
  • Scalable – process hundreds of PDFs automatically, no manual effort per document
  • Legally compliant – meets EAA 2025, BITV 2.0 and EN 301 549 simultaneously
  • For all user groups – sighted users see exactly the original, screen readers access the full structured layer
This used to be a dilemma: Either accessibility or layout fidelity – having both was barely possible. PDF/UA Hybrid combines layout fidelity and an accessible text version in one file.

🔒 Privacy & Security

How secure are my uploaded PDFs?

Comprehensive security measures:

🛡️ Technical security:
  • SSL/TLS encryption during upload
  • Secure servers in Germany (GDPR-compliant)
  • Automatic deletion after 30 days
  • No permanent storage
  • Access restrictions to your files
📋 Data protection compliance:
  • GDPR-compliant processing
  • No sharing with third parties
  • Transparent data usage
  • Opt-out possible at any time
  • Data processing agreements available
Important: Do not upload highly sensitive or personal data. For such cases we offer Secure Local Mode.
What is Secure Local Mode?

Highest data security for sensitive documents:

On-Premise Installation

PDF Accessibility Pro is installed directly in your IT infrastructure:

  • No cloud processing: All data stays with you
  • Full control: You decide who has access
  • GDPR-compliant: Ideal for public authorities and companies
  • Offline operation possible: Also works without internet
  • Customizable features: Special requirements can be implemented
Request now:

Contact us at info@barrierefrai.com for a custom offer for Secure Local Mode.

🔧 Troubleshooting

What to do when “Not enough pages available” appears?

This message appears when your monthly page quota has been used up:

🔍 Solutions:
  1. Upgrade plan: More pages with a higher plan
  2. Wait until next month: Quota renews automatically
  3. Smaller PDFs: Split the document
  4. Standard mode: Use this on the Free Trial
📊 Current limits:
  • Free Trial: 25 pages once
  • Starter: 250 pages/month
  • Pro: 600 pages/month
  • Enterprise: 1500 pages/month
“Vision AI shows empty results” – Why?

Possible causes and solutions:

🔍 Common causes:
  1. Vision AI is only available in Pro/Enterprise
  2. The PDF contains only scanned images
  3. Temporary API error
  4. The PDF file is corrupted
✅ Quick fixes:
  1. Use the “Compact” mode
  2. Check your plan status
  3. Try a smaller PDF file
  4. Contact our support

🧪 PAC Testing & Validation

What is PAC 2024 and how does it differ from veraPDF?

PAC 2024 (PDF Accessibility Checker) and veraPDF are both recognized tools for checking PDF accessibility — but they work in fundamentally different ways:

veraPDFPAC 2024
BasisISO 14289-1 (specification)Heuristics + AI
Result typePass / Fail (normative)Pass / Warning / Error
AuthorityLegally binding for PDF/UA-1Advisory / supplementary
Analysis depthTechnical structureVisual + structural

veraPDF tests precisely against the ISO standard — a 100% result means the document fully satisfies PDF/UA-1.

PAC 2024 adds heuristic and AI-powered checks: it analyses the visual layout, detects potential structural issues, and provides guidance — even when no normative violation exists.

Recommendation: Use veraPDF as the binding compliance check. PAC 2024 provides supplementary quality guidance — warnings (not errors) do not constitute standard violations.
PAC shows “Possibly inappropriate use of a Form structure element” – What does this mean?

This warning appears when PAC 2024 detects Form structure elements in a PDF’s structure tree that it cannot unambiguously map to a visible form field.

Why does it appear in PDF/UA Hybrid?

PDF/UA Hybrid has a distinctive architecture: the visible layer is the original PDF rendered as an image layer — the accessibility layer is the embedded accessible text version (PDF/UA). Form widgets from the original document are linked into the structure tree via /Form structure elements, as required by PDF/UA-1 (Matterhorn Protocol, checkpoint 14-003).

PAC sees the Form elements in the structure tree but cannot reconcile them with the visual image layer — hence the warning. This is not an error, but an inherent characteristic of the hybrid format.

✅ What you should do:

Nothing — provided veraPDF reports 100%. The structure is technically correct per ISO 14289-1.

🔍 Additional check:

Test PDF/UA Hybrid with a screen reader (NVDA + Firefox or VoiceOver + Safari). If all form fields are announced correctly, usability has been checked.

PAC AI Check: “H element detected, Score 0.78” – Do I need to act?

PAC 2024 includes an AI-powered visual analysis that classifies text in the PDF by its visual appearance. When the AI finds text that looks like a heading, it checks whether a corresponding <H> structure element exists in the structure tree.

📊 What does the score mean?
  • Score < 0.80: Low-confidence hint – no action needed
  • Score 0.80–0.89: Warning – review in context
  • Score ≥ 0.90: Manual review recommended
🔍 Why in PDF/UA Hybrid?

The visual layer consists of images of the original document. The AI “sees” rendered text in these images — the correctly tagged <H1><H6> elements exist in the embedded PDF/UA layer, but the AI cannot visually correlate the two layers.

Key point: veraPDF 100% remains the authoritative statement on PDF/UA-1 compliance. The AI score in PAC is a supplementary heuristic hint — not a standard violation.
My PDF/UA Hybrid passes veraPDF, but PAC 2024 shows warnings – Is the document accessible?

Yes — a 100% veraPDF result means full conformance with ISO 14289-1 (PDF/UA-1), the binding standard for accessible PDFs.

PAC 2024 warnings (not errors!) are heuristic advisories. They flag potential best-practice deviations but do not constitute standard violations:

  • 🔴 veraPDF FAIL = ISO violation → must be fixed
  • 🔴 PAC Error (red) = serious issue → should be fixed
  • 🟡 PAC Warning (yellow) = advisory → review in context, often unavoidable
  • 🟢 PAC Pass = no finding

PDF/UA Hybrid files can produce 0 veraPDF errors and still show individual PAC warnings. These notes must be reviewed in the context of the document.

For legal requirements (EN 301 549, EAA / European Accessibility Act, WCAG 2.1 AA): veraPDF 100% satisfies the technical accessibility requirement.
What PAC results are typical for PDF/UA Hybrid?

Due to their dual-layer architecture, PDF/UA Hybrid files can produce characteristic PAC findings:

🟡 Typical warnings (expected — no action needed):
  • “Possibly inappropriate use of a Form structure element” — appears for each embedded form field (8–20×, depending on the original)
  • “AI detected H/Table/Figure element, Score < 0.85” — AI recognition in the image layer without direct structure tree correlation
🔴 Not typical (should be investigated):
  • PAC errors (red) at structure level → missing alt text, incorrect tag order
  • veraPDF FAIL → technical error in the embedding
  • “Document is not tagged” → Hybrid generation did not complete successfully
✅ Quick verification checklist:
  • ☑ veraPDF: 100% (PDF/UA-1 Pass)
  • ☑ PAC: no red errors (yellow warnings acceptable)
  • ☑ Screen reader test: content logically readable, form fields announced
  • ☑ Document language set (e.g. de, en)
  • ☑ Document title present in metadata

Any further questions?

Contact us directly by email for individual support:

Send email: info@barrierefrai.com

Response time: 48 hours on business days