How to Create a Text Transcript That’s WCAG Conformant (with AI Help)

Text transcripts provide access to audio and/or video content for people who are deaf, hard of hearing, or prefer reading text instead of watching or listening. Here’s how to create a WCAG conformant transcript.

AI + Manual

With AI transcription tools now widely available, creating transcripts has become so much easier. Many video editors will automatically create subtitles for you (you need more than subtitles, though). You still need to review the subtitles for accuracy, but this already takes out a lot of the work.

If you have a video, you can also upload to YouTube and generate time stamped subtitles and then use those as the basis for your text transcripts.

After you have the text, you need to assign speakers to their spoken words. You can use AI tools like ChatGPT, Claude, or Gemini to help with this by giving it the initial dialogue to a speaker and then telling it to continue assigning speakers.

Other than speakers, if there is visual information being conveyed in a video (this wouldn’t be the case for a podcast), then you’d add in that visual information. For example, if there’s a PowerPoint slide on the screen showing the 10 steps to YouTube accessibility, you’d want to relay the relevant information.

Many paid AI transcription tools can identify different speakers, timestamp content automatically, and even handle multiple languages. You still need to review the output, but most of the transcription work is automated.

Accessible Transcripts

A truly accessible transcript captures more than just spoken words. It provides context, identifies speakers, and describes significant non-speech elements that contribute to understanding the content.

A complete transcript should include:

• All spoken content accurately transcribed
• Speaker identification when there are multiple speakers
• Description of relevant non-speech sounds (applause, music, significant background noises)
• Description of important visual information for video content
• Text shown on screen that isn’t spoken aloud

Format and Structure

The most common format for web transcripts is HTML, but there’s no single required design. Common approaches include:

• Simple text paragraphs with speaker identification
• Structured table format that separates audio and visual information
• Accordion-style expandable transcript that can be hidden when not needed

Simple Paragraph Format Example

The paragraph format works well for straightforward audio content like podcasts or basic interviews. Speaker changes are clearly indicated, and the flow mimics natural conversation.

Here’s an example:

Kris Rivenburgh: Welcome to our accessibility podcast.

Winston Scout: Thanks for having me today to discuss transcript creation.

Kris: Could you tell us about your experience with creating accessible content?

Winston: I’ve been working in digital accessibility for over five years, focusing primarily on making multimedia content accessible through proper transcripts and captions.

Structured Table Format Example

For more complex video content, a structured table format helps separate visual and audio information. This approach makes it easier for users to understand what’s happening visually while also following the audio content. The clear separation helps users focus on what’s most relevant to them.

Here’s an example:

Next Level Usability

A basic transcript meets accessibility requirements, but some upgrades can significantly improve the user experience.

To increase UX, consider these optional add-ons:

• Add headings to break up long transcripts into logical sections
• Include timestamps only when they provide value (for navigating to specific points)
• Add clarifying information in [brackets] to provide context
• Include hyperlinks to referenced resources
• Use formatting like bold text to emphasize important speakers or points

Best Practices for Transcripts

Creating an effective transcript involves balancing verbatim accuracy with readability. While capturing all important information is crucial, a transcript that’s difficult to read defeats its purpose. These best practices help with balance:

1. Prioritize accuracy over verbatim transcription – leave out filler words (“um,” “uh”)
2. Format for readability with proper paragraphs and spacing
3. Include only relevant visual information – focus on what’s necessary for understanding
4. Use proper punctuation to maintain the meaning and flow of speech
5. Consider your audience – technical content may require more detailed descriptions

Where to Put Transcripts

The placement of your transcript can significantly impact its discoverability and usefulness.

• On the same page as the media, typically below the player
• On a separate linked page (clearly labeled as “Transcript”)
• Within an expandable/collapsible section on the same page
• As a downloadable file (though an HTML version should still be available)

Note that our transcript hosting service, TranscriptHost.com was specifically created so everyone can easily upload transcripts to a centralized hosting location and then simply link to them in descriptions.

Summary

An accessible transcript is more than just copying and pasting text. By including speaker identification, visual information, proper formatting, and enhanced navigation, you not only make your audio and video more accessible, but you also unlock big-time marketing and usability benefits.

If you don’t know where to put your transcript, try TranscriptHost.com.

Related Posts

• How to Use ChatGPT to Create an Accessible Text Transcript
• Link to Your Text Transcript from Our Third-Party Hosting Platform
• How to Make YouTube Videos ADA Compliant
• Do YouTube Videos Need Text Transcripts for Accessibility?