Using Cloudinary MediaFlows for Accessibility

Alt text is essential. Without it, screen readers can’t interpret images, leaving key content inaccessible to the estimated 43.3 million people who are blind and 295 million who have moderate to severe vision impairment, roughly 4% of the world’s population. 

Accessibility is not just a moral imperative; it’s also a legal and commercial one. The European Accessibility Act, in force since June 28, 2025, applies to a wide range of products and digital services and mandates meaningful text alternatives for non-text content, in line with WCAG standards. Failing to comply can limit your market access and trigger enforcement.

Done well, alt text does more than improve accessibility. Clear, accurate descriptions enhance overall user experience, extend your product’s reach, and even strengthen SEO, as search engines read alt text, too.

The European Accessibility Act requires a text alternative for all non-text content, including images, icons, time-based media, and more, to ensure the same information is accessible to assistive technologies. 

• Meaningful vs. decorative. Provide clear, intent-equivalent alt text for informative images so users get the full context. If an image is purely decorative, use an empty alt attribute so screen readers skip them, reducing unnecessary noise for users who rely on assistive tech. 
• Screen readers and pronunciation. If your site offers translated or localized content, mark language correctly so speech engines and braille displays switch pronunciation rules appropriately. 
• Global audiences need local text. For accuracy and comprehension, alt text should always appear in the viewer’s language. But maintaining manual consistency across large content libraries and multiple locales is rarely scalable; automation or translation workflows can help close that gap.

Even with the best intentions, many teams face practical challenges when implementing accessible alt text at scale. The most common pain points include:

• Volume and velocity. Campaigns produce thousands of visual assets across markets.  Manual captioning processes simply can’t keep up with the speed and volume of content creation, and translations make it even more complex.
• Consistency. When many contributors are involved, style, tone, and level of detail often vary. This inconsistency leads to uneven quality and can undermine the overall accessibility experience.
• Governance. Keeping descriptions attached to assets as metadata and ensuring updates flow through every version and locale is difficult without automation, structured processes, and clear ownership.

Creating accessible, localized content at scale requires a repeatable workflow that combines automation and accuracy. Cloudinary MediaFlows, a workflow automation engine for visual media, enables teams to build and scale no-code workflows to manage and process assets in an intuitive interface. 

With MediaFlows, you can automatically trigger AI to generate descriptive alt text, and then translate it for global audiences, store it as contextual metadata to support both assistive technologies and SEO, and publish consistently across channels. 

Here’s an automation you can build with Cloudinary MediaFlows:

1. Upload asset and trigger workflow. Begin the workflow whenever a new image is uploaded to the product environment.
2. Analyze and caption images. Use AI to analyze an image and automatically generate a concise, descriptive text alternative. 
3. Enrich metadata. Store the original description and translations to the appropriate contextual metadata fields, ensuring it remains associated with the asset throughout its lifecycle.
4. Generate multilingual alt text. Automatically create localized versions of the alt text with Google Translate or other tools, to support global audiences.
5. Route assets. Write back the translations, tag the asset as “accessibility-ready,” and route it to the correct folder, collection, or publishing step for deployment.

Tip: If you also need to update older assets, Cloudinary’s blog includes guidance on backfilling alt text for existing library images using MediaFlows and AI Content Analysis.

Check our documentation for a quick tutorial that shows how this can be done in MediaFlows, and a more detailed, step-by-step explanation showing how the relevant PowerFlows are built:

Accessibility can be easily complied with and used to create better experiences with media, even for people with disabilities. To do this at scale, automated flows across media assets, their descriptions, metadata, alt text and more should be executed, triggering AI through automation and serving it all within websites and applications.

Cloudinary’s MediaFlows can be a great way to do this. Using it you can take media assets and make them accessibility ready, without manual work and error-free. Try it here.