How AI-Powered Audio Descriptions Are Changing Higher Education

While captions and transcripts have become standard practice, audio descriptions for learners with visual impairments often remain overlooked, despite visual impairments being among the top 10 disabilities among adults, according to the CDC.

What are Audio Descriptions? 

Audio descriptions provide verbal narrations of essential visual elements within educational videos. These descriptions work alongside dialogue to communicate actions, on-screen text, visual demonstrations, gestures, and other details to students who are blind or have low vision.

For example, consider an instructional video where an instructor points to a specific section of a diagram. An effective audio description might state: “The instructor highlights the mitochondria section in the cellular diagram, located in the upper right quadrant.”

This level of detail helps ensure all students receive the same information, which supports both compliance with accessibility standards like WCAG 2.1 and ADA Title II, while providing equitable learning experiences.

Barriers to Implementation 

Despite their value, audio descriptions remain underused across higher education, primarily because the traditional creation process is time and resource-intensive. Creating high-quality descriptions typically requires:

• Writing a script based on the visual content
• Recording the narration
• Editing the video to fit the descriptions without interrupting the original audio

For instructors managing multiple courses or institutions with limited accessibility resources, these requirements often make comprehensive audio description implementation seem unattainable.

How AI is Improving Audio Descriptions

Artificial intelligence technology is addressing these traditional barriers by changing how institutions approach creating audio descriptions. AI can now analyze a video’s visual content, identify key elements, and generate descriptive narration quickly and at scale. 

Some ed-tech solutions, like the YuJa Enterprise Video Platform, have introduced tools that allow users to generate enhanced, pause-and-play audio descriptions using AI. Instead of layering narration over the original video, this approach pauses the video, plays the description, and then resumes. This helps ensure nothing is missed and avoids overlapping or confusing audio. 

While human review is still important, AI can provide a strong starting point, simplifying offering accessible videos across entire libraries.

Using AI to Scale Accessibility

As educational technology continues to evolve, accessibility features will become more sophisticated and integrated into teaching and learning. Putting new technologies to use will help institutions shift from reactive accommodation to proactive inclusive design, as they embed capabilities into content creation workflows. 

By staying informed about new tools and best practices, institutions can take meaningful steps toward more inclusive learning environments while keeping pace with changing expectations around digital equity and compliance.