March 2026: Audio Description Tracks in Streaming Manifests - Actions Required - Updated Notice – Swank Motion Pictures, Inc.

This is a follow-up to our original notice. The streaming manifests referenced below have been updated. Please review the changes carefully.

Overview

Swank Cloud Streaming has introduced updates to streaming manifests that include audio description (AD) tracks. This change supports our WCAG 2.1 AA compliance initiative under federal accessibility requirements, with full compliance required by April 24, 2026.

As of March 2026, streaming manifests in Production may now include both a main audio track and an audio description track for supported titles. Partners with custom players need to support these updated manifests to avoid user confusion or playback issues.

If you only use the Swank Cloud Streaming Portal UI and our integrated player, no changes are required. Our Player work is completed and you'll see AD tracks as the come available.

Currently, only a limited number of titles have audio description available from our represented Studios. This will grow over time as additional audio tracks become available and are processed into our Cloud Platform. Please reach out to your Account Representative for more information on availability.

What's Changed

Partners consuming our API for streaming content links will now see two audio tracks for the same language (for example, en-US).

DASH Manifest Example (Windows / Android)

<!-- Main English audio -->
<AdaptationSet mimeType="audio/mp4" lang="en-US">
  <Role schemeIdUri="urn:mpeg:dash:role:2011" value="main"/>
</AdaptationSet>

<!-- English descriptive audio -->
<AdaptationSet mimeType="audio/mp4" lang="en-US">
  <Role schemeIdUri="urn:mpeg:dash:role:2011" value="description"/>
  <Accessibility schemeIdUri="urn:tva:metadata:cs:AudioPurposeCS:2007" value="1"/>
</AdaptationSet>
```

Key takeaways:

lang="en-US" – Both sets are identified as English (U.S.).
<Role … value="main"/> – Identifies the primary audio track.
<Role … value="description"/> – Explicitly identifies the audio description track.

HLS Manifest Example (Apple devices)

#EXT-X-MEDIA:TYPE=AUDIO,GROUP-ID="audio/mp4a.40.2",LANGUAGE="en-US",
NAME="English",AUTOSELECT=YES,DEFAULT=YES,CHANNELS="2",
URI="audio-en-US.m3u8"

#EXT-X-MEDIA:TYPE=AUDIO,GROUP-ID="audio/mp4a.40.2",LANGUAGE="en-US",
NAME="English (Descriptive)",AUTOSELECT=NO,DEFAULT=NO,CHANNELS="2",
URI="audio-en-US-x-ad.m3u8",
CHARACTERISTICS="public.accessibility.describes-video"

Key takeaways:

LANGUAGE="en-US" – The descriptive track is still identified as English (U.S.).
NAME="English (Descriptive)" – Provides a clear, human-readable label for the player UI.
CHARACTERISTICS="public.accessibility.describes-video" – The machine-readable flag indicating the track's accessibility purpose.

Action Required

The April 24, 2026 deadline is now less than 30 days away. Cloud Streaming partners with their own video players should complete the following as soon as possible:

Review and test integration with the updated manifest structure.
Verify playback across all relevant platforms (Windows, Android, Apple TV / iOS).
Confirm that audio description appears correctly in your player's language or audio selection menu.
Validate that no duplicate or unlabeled audio tracks appear for the same language.

Failure to implement these updates may result in duplicate or misidentified audio tracks and inconsistent playback behavior.

If you only use the Swank Cloud Streaming Portal UI and our integrated player, no changes are required.

Why This Matters

Audio description ensures that users who are blind or have low vision receive a complete and inclusive viewing experience. Adopting this enhancement aligns your applications with WCAG 2.1 AA success criteria 1.2.3 and 1.2.5 and supports federal accessibility requirements under 28 CFR § 35.200, effective April 24, 2026.

Expected Player Behavior

If your application uses a custom video player, your player must distinguish between the primary audio track and the audio description track, even when both tracks use the same language code, such as en-US. Audio description should appear as a separate option in the player’s audio track selection UI, with clear user-facing labeling.

Players should not rely on language code alone when identifying or grouping audio tracks. Instead, they should use the manifest metadata provided for each format:

DASH: use the track role metadata, including main and description
HLS: use the track characteristics metadata, including public.accessibility.describes-video

Where the player supports display labels, the UI should present clearly differentiated options such as:

English
English – Audio Description

The primary audio track should remain the default playback selection unless the end user explicitly chooses the audio description track. This helps ensure a consistent playback experience while still exposing accessibility features appropriately.

What to Test

Please validate your implementation using titles that include both a primary audio track and an audio description track for the same language.

Your testing should confirm the following:

The player displays both audio tracks as separate choices in the audio track menu
The player clearly labels the audio description track for end users
The player does not show two indistinguishable entries such as two tracks both labeled only “English”
The primary audio track remains the default selection unless the user chooses otherwise
The audio description track can be selected manually and plays correctly
The player behaves correctly across all supported target platforms
Any saved language or playback preferences do not cause the wrong same-language audio track to be selected automatically

You should also confirm that titles without audio description continue to behave normally and do not require any special handling. The updated manifest structure should be handled gracefully whether one or multiple audio tracks are present for a language.

Common Implementation Pitfalls

We recommend reviewing your player logic for the following common issues:

Grouping audio tracks by language code only
Displaying multiple same-language tracks without distinct labels
Failing to expose the audio description track in the audio selection menu
Automatically selecting the audio description track when it should be user-selected
Ignoring DASH role metadata or HLS accessibility characteristics when building the audio track menu

If your current implementation assumes there will only ever be one audio track per language, updates will likely be required. As noted in this notice, partners consuming Swank streaming links may now receive two audio tracks for the same language.