Some events are broadcast as live audio, such as webcasts, speeches, or radio streams. To make this content accessible, a text equivalent must be provided. A text equivalent is a live, readable text version of the spoken audio that conveys the same information as it is being delivered.
For radio streams, this can be provided through live captions or a continuously updated live transcript displayed on a website or companion page. The text should reflect the spoken content in real time, allowing people who are deaf, hard of hearing, or unable to process audio to follow what’s being said as the broadcast happens.
WCAG Success Criterion
1.2.9 Audio-Only (Live) – Level AAA
An alternative for time-based media that presents equivalent information for live audio-only content should be provided.
Live audio content must include a text alternative that presents equivalent information with minimal delay.
Why This Matters
People who are deaf, hard of hearing, or cannot process auditory information rely on text to understand live content.
A live text alternative ensures they can participate and stay informed as the event unfolds.
For example:
- During a live press conference, captions (text displayed in real time for spoken audio) or a live transcript (a running text record of what is being said) help users read what’s being said.
- In a live radio broadcast or online audio stream, providing a text version ensures everyone can access the same information.
When You Should Use It
You should provide a live text alternative whenever your content includes:
- Live audio-only broadcasts such as webcasts, radio streams, or audio announcements.
- Spoken events like live speeches, conferences, or podcasts.
This does not apply to private or two-way communication (e.g., phone calls or web meetings). It focuses on broadcast or public streaming, where users listen but cannot directly speak.
How to Provide a Text Alternative
There are several ways to make live audio-only content accessible. The method you choose must ensure that spoken information is available in text as it is delivered.
Live Captioning (Preferred Method)
Live captioning provides real-time text that closely follows the live audio. This does not require a professional captioning service in all cases. However, professional human captioners generally provide the highest accuracy, especially for complex, fast-paced, or technical content.
Captions typically appear with a short delay and should include descriptions of important sounds when relevant.
Automated (AI-generated) live captions may be used, but their accuracy varies. They are more suitable for simple, clear speech and low-risk content. If automated captions contain frequent errors, they may not meet accessibility requirements.
Link to a Prepared Script or Transcript
If a live event follows a prepared script (such as a formal speech or reading), you may provide a link to that script only if the spoken content closely matches the written text.
If the speaker deviates from the script, adds explanations, or answers questions, the script alone is no longer sufficient. In such cases, a live text alternative (such as captions or a live transcript) is still required to meet accessibility requirements.
Text-Based Live Feed
Provide a dedicated text feed or section on your webpage where typed updates reflect the spoken content in real time.
For example, an embedded chat or scrolling text window may be used to display live transcriptions or summaries. This method is acceptable only if the updates accurately and continuously reflect what is being said.
Integrated Captioning in the Media Stream
Use streaming technologies (such as HTML5 media players or broadcast systems that support live captions) to embed captions directly into the live audio or video stream.
This allows users to enable captions within the player while accessing the live content.
Examples
- A public relations firm hosts a live product launch. They hire a captioning service whose live text output appears below the audio stream on the same webpage.
- A theatre group broadcasts a live radio play that follows a written script. They provide a downloadable link to the full script for real-time reading.
- A news company streams a CEO’s live statement, and since it is a prepared speech, they publish the statement text on the same page while streaming the audio.
Common Mistakes
- No live captions or transcript
Providing only a post-event transcript is not sufficient. Users who cannot hear need access to the information as it is delivered. Without a live text alternative, they miss all meaningful content during the event. - Inaccurate or delayed captions
Live captions should appear within a few seconds of the spoken audio. Delays of tens of seconds or minutes, or captions that skip key information, fail to meet accessibility requirements and prevent real-time understanding. - Using untrained operators or low-accuracy systems
Live transcription must be accurate and readable. Human captioners generally provide the highest quality, especially for complex or fast-paced content. Automated speech recognition (ASR) can be used for simple content, but if accuracy is poor or errors are frequent, it does not meet accessibility standards.
Good Practices
When implementing live text alternatives:
- Ensure minimal delay between speech and text output.
- Include notes for non-speech sounds that convey meaning (e.g., “[applause]”, “[music playing]”).
- Clearly identify the speaker in multi-person events.
- Provide the text in a high-contrast, readable format on the same page as the audio player.
- Save the transcript afterwards so users can review it later.
Wrapping Up
Providing a live text equivalent makes audio-only broadcasts accessible to everyone.
Whether through live captions or a shared script, users who cannot hear can still follow and understand the content as it happens.
