SiteLint documentation and guidelines

Verify audio-only (live)

Description

This rule does not automatically validate anything. It simply reminds you to manually verify that live audio-only content has a text alternative (such as real-time captions or a transcript) available to users. Live audio includes web-based conferencing, live radio streams, breaking-news audio feeds, and any other real-time audio that is broadcast over the web with no more than a broadcast delay.

WCAG Success Criterion 1.2.9 Audio-only (Live) (Level AAA) requires that an alternative for time-based media presenting equivalent information is provided for live audio-only content. Unlike prerecorded audio, where captions or transcripts can be prepared in advance, live audio demands a real-time solution such as a stenographer, a live captioning service, or a prepared script if the speaker follows one. This rule flags the page for a human reviewer to confirm that a suitable text alternative exists and is accessible.

Disabilities impacted

  • Hearing (deaf, hard of hearing) – Users who are deaf or hard of hearing cannot access information conveyed solely through live audio. Real-time captions or a live transcript are essential for these users to participate in web conferences, follow live announcements, or consume streaming audio content. Without a text alternative, the entire broadcast is inaccessible.
  • Screen reader users – Users who rely on screen readers may have difficulty navigating audio-only content, especially when no visual interface or text track is available. A text alternative allows screen readers to convey the live audio information in a format the user can process at their own pace.
  • Cognitive (attention deficits, memory impairments) – Users with cognitive disabilities may struggle to process fast-paced live audio, retain information heard only once, or filter out background noise. A text alternative provides a persistent reference that can be re-read, supporting comprehension and reducing cognitive overload.

Why it matters

Live audio is increasingly used for critical communications: emergency alerts, earnings calls, press conferences, classroom lectures, and live-streamed events. Excluding deaf or hard-of-hearing users from real-time information is not only a compliance risk under ADA, EAA, and Section 508 but also a significant equity issue. WCAG 1.2.9 (Level AAA) sets the expectation that content providers plan for live accessibility in the same way they plan for broadcast infrastructure. Failing to provide a text alternative means that a segment of the audience is locked out of time-sensitive content. Beyond legal compliance, providing live captions or transcripts improves the experience for all users: non-native speakers, users in noisy environments, and anyone who prefers reading over listening can all benefit.

Coding problems and solutions

Common coding problems

Because this rule is a manual reminder, the items below represent what a human reviewer should look for during an audit of live audio-only content.

  • No live captioning service integrated – The page embeds a streaming <audio> element but does not include a real-time captioning widget, an embedded stenographer feed, or any mechanism that delivers text synchronized with the live audio.
  • Transcript not available or not linked – For scripted live audio (e.g., a CEO reading a prepared statement), the prepared transcript is not linked from the same page or is linked in a location that screen readers cannot easily discover.
  • Captions not visible or not keyboard accessible – A live caption feed is present but rendered in a way that is not perceivable (e.g., off-screen, hidden behind a non-focusable toggle) or cannot be navigated using a keyboard.
  • Auto-generated captions without human review – The page relies entirely on automatic speech recognition without a human operator, producing captions with high error rates for domain-specific terminology, multiple speakers, or accented speech.
  • No indication that a text alternative exists – The live audio player is presented without any visible label or link informing users that captions or a transcript are available, making the text alternative effectively undiscoverable.

How to fix it

Integrate a live captioning service

Use a professional live captioning provider that generates real-time captions via a stenographer or speech-to-text operator. Embed the caption output into the page using an accessible container that updates dynamically.

Correct example
<div aria-live="polite" aria-atomic="true" aria-label="Live captions">
  <h3>Live captions</h3>
  <p id="caption-feed">
    Captions will appear here during the broadcast.
  </p>
</div>
Incorrect example
<!-- No caption container at all -->
<audio src="live-broadcast.mp3" controls></audio>

Provide a transcript for scripted live audio

When the speaker follows a prepared script, a transcript can serve as the text alternative. Link the transcript prominently beside or beneath the audio player so it is immediately discoverable.

Correct example
<audio src="press-release-live.mp3" controls></audio>
<p>
  <a href="/transcripts/press-release-may-2026.pdf">
    Read the prepared transcript of this live press release
  </a>
</p>
Incorrect example
<!-- Transcript exists but is not linked from the audio page -->
<audio src="press-release-live.mp3" controls></audio>
<!-- Users must discover the transcript URL on their own -->

Update captions dynamically using JavaScript

When consuming a live captioning API, update the caption container with new text as it arrives. Use an aria-live region so assistive technologies announce changes automatically. Always handle connection errors gracefully.

Correct example
const captionFeed = globalThis.document.getElementById('caption-feed');
const captionSocket = new WebSocket('wss://captions.example.com/live');

captionSocket.addEventListener('message', (event) => {
  try {
    const data = JSON.parse(event.data);
    if (typeof data.text === 'string' && data.text.length > 0) {
      captionFeed.textContent = data.text;
    }
  } catch {
    console.error('Failed to parse caption data');
  }
});

captionSocket.addEventListener('error', () => {
  captionFeed.textContent =
    'Caption feed interrupted. Please refresh the page.';
});
Incorrect example
// Using a string in setInterval (implies eval) and no error handling
const captionFeed = document.getElementById('caption-feed');
setInterval("captionFeed.innerText = 'static placeholder'", 5000);

Make captions visible and keyboard accessible

Ensure the caption area has sufficient color contrast, a readable font size, and can be focused or scrolled using the keyboard. Avoid placing captions in a non-scrollable container that clips long passages.

Correct example
<div
  aria-live="polite"
  tabindex="0"
  role="region"
  aria-label="Live caption feed"
  style="
    background: #fff;
    color: #000;
    border: 2px solid #333;
    padding: 1rem;
    max-height: 200px;
    overflow-y: auto;
    font-size: 1.125rem;
    line-height: 1.5;
  "
>
  <p id="caption-feed">Waiting for captions…</p>
</div>
Incorrect example
<!-- Captions hidden off-screen and not focusable -->
<div style="position: absolute; left: -9999px;" aria-hidden="true">
  <p>Live captions would appear here.</p>
</div>

Announce the availability of the text alternative

Place a visible label or link near the audio player that tells all users that captions or a transcript are available. Use plain language so the instruction is understandable regardless of assistive technology.

Correct example
<audio src="live-townhall.mp3" controls></audio>
<p>
  <button type="button" aria-expanded="false"
          aria-controls="caption-panel">
    Show live captions (English)
  </button>
</p>
Incorrect example
<!-- Audio player with no mention of captions at all -->
<audio src="live-townhall.mp3" controls>
  Your browser does not support the audio element.
</audio>

Known Limitations

  • Single-page applications (SPAs) – Live audio streams and captioning widgets injected dynamically after the initial page load will not be detectable if the DOM changes after the audit runs. A human reviewer must inspect the rendered page in its fully interactive state.
  • Shadow DOM – <audio> elements and caption containers inside closed shadow roots are not traversable by the rule’s selector. Open shadow roots are visible but require the audit to support shadowRoot traversal, which may not cover all nesting levels.
  • Manual verification required – The rule does not automatically validate anything. It only reports a reminder when <audio> elements are found. Whether the text alternative is actually present, accurate, and accessible depends entirely on the reviewer’s judgment.
  • Cross-origin iframes – Live audio or captioning content embedded via cross-origin <iframe> elements is inaccessible to the audit script due to same-origin policy. The reviewer must manually inspect the embedded content.
  • Real-time accuracy cannot be automated – No automated tool can verify that live captions are semantically accurate, properly synced, or cover the full duration of the broadcast. A human must observe the live experience to confirm adequacy.

Resources

Rule

  • Audit: accessibility
  • Standard: WCAG
  • Level: AAA
  • Success Criteria: 1.2.9
  • Rule ID: audio-only-live