How it works

A simple link. A serious engine behind it.

To your users, Visionably is one extra link beside the PDF. To your auditor, it's a complete accessibility pipeline producing standards-validated documents on demand. This page explains what happens between the click and the download.

01 · Understanding the document

We read the document
the way a person would.

Most accessibility tools treat a PDF as a flat bag of text. Visionably looks at the actual page — the same picture you'd see — and works out what every block of content is for. Is this a heading? A footnote? The caption under a chart? A column that wraps to the next page?

Once we know the structure, we know the reading order. Screen readers follow that order, instead of jumping around the page randomly.

What we identify

  • · Headings, sub-headings, section hierarchy
  • · Reading order across multi-column layouts
  • · Tables, including row/column headers and spans
  • · Lists, captions, footnotes, sidebars
  • · Charts, diagrams, figures with their captions
  • · Headers, footers and page numbers (so we can hide them)
  • · Equations, formulas and inline math
  • · Forms, signatures, fillable fields

Generated alt text

"Stacked bar chart comparing revenue by business segment over five fiscal years. Services revenue rises from ₹120 crore in FY21 to ₹340 crore in FY25, overtaking Hardware in FY24. Hardware remains roughly flat between ₹180 and ₹210 crore. Subscriptions, introduced in FY23, grow from ₹40 to ₹95 crore."

Reading time

~ 18 seconds spoken

Length

61 words

02 · Describing what's there

Charts described.
Tables rebuilt. Math read.

A chart isn't accessible if its alt text says "Figure 3." A table isn't accessible if a screen reader can't tell which cells are headers. An equation isn't accessible if it gets spoken as "x squared plus blah blah."

For every figure, we write a real description grounded in the surrounding text. For every table, we restore the header relationships. For every equation, we emit both the spoken form and the underlying math.

03 · Proving it's compliant

Every file is checked
before it leaves us.

Compliance isn't a button you press. It's something you prove. Every document Visionably serves is automatically validated against the four accessibility standards your auditor will quote — and we attach a machine-readable conformance report that maps each finding back to the specific WCAG success criterion, so your legal team has a clear paper trail.

WCAG 2.1 AA

Validated · web standard

PDF/UA-1

Validated · ISO 14289-1

Section 508

Validated · US federal

RPwD Act

India · 2016

For developers

Trigger conversions
in bulk. Or one at a time.

Most customers just paste the link snippet onto their portal and never write a line of code. If you want to script bulk conversions — say, for a content migration or a nightly index — the API is here.

Drop-in link html 
<!-- next to every PDF on your portal -->
<a href="https://visionably.ai/v/{your-key}?src={pdf-url}">
  Accessible version
</a>
Bulk API curl 
$ curl -X POST https://api.visionably.ai/convert \
    -H "Authorization: Bearer $KEY" \
    -d '{"url":"https://acme.in/doc.pdf"}'

{ "status": "ready", "url": "..." }

Full API reference shared with customers on request. dev@visionably.ai

See it on one of your documents.

Book a demo