Audio descriptions make video content accessible to students who are blind or have low vision by describing important visual information that isn’t conveyed through dialogue or sound alone. These descriptions allow students to fully engage with course materials, especially when visual elements are critical to understanding.
Providing audio descriptions meets WCAG 2.1 AA Success Criterion 1.2.5 (Audio Description – Prerecorded) and is a key part of equitable course design for students with visual disabilities.
AI tools like ChatGPT, Claude, and Gemini can help generate audio description scripts when you provide them with detailed prompts and video context. While these tools don’t create the audio files themselves, they can assist you in writing the descriptive narration.
I am creating an audio description for a video used in my online course. Please write a descriptive script that narrates the important visual information for students who are blind. The video is 2 minutes long and includes no dialogue. It shows a science experiment where a beaker of clear liquid changes color, foams up, and overflows. Include time stamps and ensure your description is clear, concise, and follows WCAG 2.1 AA guidelines.
AI-Generated Description Script:
[00:00–00:15] A glass beaker filled with clear liquid sits on a metal lab table.
[00:16–00:35] A hand wearing blue gloves adds powder into the beaker using a silver scoop.
[00:36–00:50] The liquid changes color from clear to bright orange and bubbles.
[00:51–01:10] Foam rises quickly and spills over the edge of the beaker onto the table.
[01:11–02:00] The reaction slows, leaving orange foam covering the table while the liquid inside the beaker settles.
Narrate only what’s visually important—don’t describe every detail.
Match the tone and pace of the original video.
Focus on actions, facial expressions, text on screen, and scene changes.
Use time stamps to align with key visual moments.
If you prefer AI narration, you can record your script using text-to-speech (TTS) tools like Murf.ai, Play.ht, or Microsoft Copilot.
Select a short instructional video or demonstration from your course. Mute the sound and play the video. As you watch, write down what a blind student would need to know to understand the visuals. Then use ChatGPT with the prompt above to generate an audio description script. Compare your version with the AI’s and revise as needed.
Try recording the description and layering it over the muted video using your LMS, Panopto, or another tool.
If you have questions about writing or recording audio descriptions, or need feedback on your script, contact Instructional Accessibility Designer Terisa O’Dowd at:
📧 todowd1@twu.edu