Describewright – AI Audio Description for Accessible Video

Description

Audio description for every WordPress video, written by AI and checked by you.

Blind and low-vision viewers hear what is said in a video, but not what is shown: the slide with the new opening hours, the name under the speaker, the map, the chart, the person who walks in. Audio description fills that gap with short spoken descriptions in the pauses of the speech. WCAG 2.1 requires it for prerecorded video at Level AA (success criterion 1.2.5).

Describewright writes those descriptions for you:

1. Open Media → Audio description and click Describe next to a video.
2. Describewright finds the pauses in the speech, spots the scene changes and picks still frames.
3. An AI model that can see images writes one description per pause, sized to fit it, including the on-screen text.
4. You check each description next to the playing video, change what you like, and save.
5. Your videos get an Audio description button. Viewers turn it on and hear each description at the right moment, and screen-reader users get it announced.

Why now

Audio description is part of WCAG 2.1 Level AA, the standard behind:

• ADA Title II (US): state and local governments, public schools, colleges and universities must meet WCAG 2.1 AA on their websites from April 26, 2027 (50,000 residents or more) and April 26, 2028 (smaller entities and special districts).
• The European Accessibility Act, which has applied across the EU since June 28, 2025, and the EU Web Accessibility Directive for public-sector websites (both through EN 301 549).
• Section 508 for US federal agencies (WCAG 2.0 AA, which includes the same success criterion).

Captions alone do not meet 1.2.5. Professional audio description is usually priced per minute of video. Describewright uses the AI provider you set up in WordPress, so you pay that provider directly, with no per-minute fee from us.

Features

• Finds the pauses in the speech by listening to the sound track, or uses your captions when you have them from Captionwright.
• Sees the video: detects scene changes and sends still frames from the right moments to the AI.
• Fits the pause: each description gets a word budget from the length of its pause, and the editor warns when one is too long.
• Reads on-screen text word for word: titles, names, slides, signs and numbers.
• Notes for the AI: tell it the names and context once (“Mayor Ana Ruiz speaks at the podium”) and it uses them.
• Editor with frames: every description shows the frame it describes, with editable start and end times and a live read-aloud preview.
• Audio description button in the core Video block and the [video] shortcode, with a language picker when you have several languages, keyboard and screen-reader friendly.
• Description text under the video: while a description is read, it also appears as text below the player for viewers who follow along by sight. You can turn this off.
• Described transcript block and [describewright_transcript id="123"] shortcode: the speech and the descriptions as text, in time order (WCAG 1.2.3 and 1.2.8). Click a time to jump the video there.
• Standard WebVTT description files (kind="descriptions") in your uploads folder. They are yours to keep.
• Built on WordPress AI: Describewright uses the AI provider you set up once under Settings → Connectors (WordPress 7.0+). It has no API key or AI settings of its own.

Who it is for

Cities, counties and public agencies that publish council meetings and announcements; schools, colleges and universities with lecture and campus videos; course creators and membership sites; museums, libraries and nonprofits; and agencies that look after accessibility for their clients.

Describewright Pro

• The whole library in one queue: describe every video without audio description, with a list of videos to retry if one fails.
• Extended audio description: the player waits while a long description is read, so nothing important has to be left out.
• Translate descriptions into 30 languages with the same timing; viewers pick the language next to the button.

External services

Describewright does not connect to any AI service itself. It asks WordPress’ AI client (WordPress 7.0+), which sends the request to the AI provider you set up under Settings → Connectors, with the API key you entered there. This happens only when you start one of the actions below, and nothing is sent to Karelia Labs.

• Create audio description sends still frames from the video (JPEG, up to 512 pixels wide, each labelled with its time), the video’s title, your notes for that video, the pauses (start and end times and a word limit), the language and style you chose, up to four descriptions written earlier in the same video (so the AI does not repeat them) and, if Captionwright has captions for the video, the words spoken in and around that part of the video, with their times.
• Test the connection sends one built-in 32×32 red test image and a fixed question. Nothing from your site.

The provider you choose in Connectors handles the data under its own terms and privacy policy.