This translation has not been editorially reviewed yet. The German version is authoritative. Deutsche Fassung
Accessibility
How does a live audio stream become accessible?
A radio stream, a conference call, a live reading: real-time audio only does not reach deaf and hard of hearing people. Level AAA criterion 1.2.9 requires an equivalent text alternative that reproduces the live content. In practice, this is a live transcript by written interpretation or, in the case of prepared content, the script, which is published in parallel.
What the Criterion Requires
For audio-only content sent live, there is a text version that renders the content equivalent. Equivalent means the same information content, in reasonable delay. For a text read aloud, the script, published at the time of transmission, is sufficient. Free conversation requires a transcript by humans (interpretation) or automatic recognition that someone corrects during the broadcast.
Why this is at level AAA
Written interpretation costs one trained person per hour of broadcasting time; For current programs, this adds up. The WCAG therefore assigns live audio to the level AAA, while live video with subtitles (1.2.4) is already on AA. For organizers with individual live appointments, the criterion is easily achievable: one transcript per appointment, announced and subsequently published as text.
This is how we deal with it
Live content cannot be assessed in a source code measurement; the test service finds audio elements and stream embedments and identifies them. Whether a text alternative was available during the broadcast, the organizer proves. The criterion belongs to the category AAA in knowledge.
Examples
Live radio without text version
<audio controls src="https://stream.example/live"></audio><audio controls src="https://stream.example/live"></audio>
<section aria-live="polite" aria-label="Live-Mitschrift">
<p>19:02 Moderatorin: Guten Abend, heute sprechen wir über …</p>
</section>The transcript is created by written interpretation or corrected speech recognition and runs alongside the stream.
- Plan for each live broadcast, whether a script is available (then publish) or freely spoken (then book written interpretation).
- Show the transcription on the same page as the stream, in a text area that updates itself and is readable with read-out programs.
- Announce the text alternative in the invitation so that those affected know that they can participate.
- Provide the transcript permanently after the shipment as a transcript.
