The short answer
Board meetings are the biggest single kind of video on most district channels. In our scan of 17 district channels, livestream replays and board meetings made up most of the uploads, and almost none had real captions. Under the ADA Title II web rule, a meeting video you post after your deadline has to meet WCAG 2.1 AA. For a meeting, that means three things:
- While it's live: captions on the stream (WCAG 1.2.4).
- The replay: accurate captions, with who is speaking (WCAG 1.2.2).
- If the picture carries information the audio doesn't, such as unread slides or a hand vote: audio description on the replay (WCAG 1.2.5).
Live stream vs replay
A livestream and its replay are treated as two different things. While the meeting is live, WCAG 1.2.4 asks for captions on the live audio. DOJ applied the same compliance dates to live captions as to everything else and declined to set an accuracy number or method, leaving it to each district to pick “the best way to comply with this success criterion based on current technology.” DOJ also noted W3C's guidance that automatic captions aren't sufficient on their own unless confirmed to be fully accurate.
Once the stream ends and the video stays on your channel, it's recorded video. Recorded video falls under 1.2.2, which asks for captions that are accurate, and under 1.2.5 for audio description. YouTube's live auto-captions don't carry over: YouTube Help says live automatic captions don't remain on the video after the stream ends and new automatic captions are generated instead. So the replay needs its own accurate track. The livestream captions guide covers options for the live part, from YouTube's auto-captions to a human captioner (CART).
Speaker names matter more here than anywhere
W3C's intent for both caption criteria says captions “identify who is speaking.” In most videos the picture makes that obvious. In a board meeting it usually doesn't: a fixed wide camera, nine people at a dais, and a voice saying “I'll second that.” A caption track that never names speakers can't be followed. Good meeting captions:
- Name each speaker when they start, with the title people use in the room: “PRESIDENT GARCIA:”, “TRUSTEE OKAFOR:”, “SUPERINTENDENT LEE:”.
- Name public commenters as the chair calls them: “SPEAKER (MARIA ALVAREZ):”.
- Get the numbers right. Motion numbers, dollar amounts and vote counts are the record: “Resolution 26-114”, “$4.2 million”, “5 to 2”.
- Mark the sounds that carry meaning: [gavel], [applause], [crosstalk], [inaudible], [microphone off].
Votes by show of hands, slides and other things only on screen
Audio description (WCAG 1.2.5) is needed when the picture shows important information the soundtrack doesn't say. In board meetings, the usual cases are:
| What happens on screen | Needs description? | How to avoid it next time |
|---|---|---|
| Vote by show of hands, result never said aloud | Yes. Who voted how is the record. | Chair announces the count and the no votes by name |
| Budget slides nobody reads aloud | Yes, the key figures on them | Presenter reads the headline numbers |
| Name and title captions on screen only | Yes, if speakers aren't introduced | Chair introduces each presenter |
| “As you can see on this map…” | Yes, what the map shows | Say what the map shows |
| Everything is said out loud | No. Captions are enough. | Nothing to change |
Meeting minutes don't replace captions
A common question: if the minutes are posted, isn't the video covered? No. WCAG 1.2.2's only exception is video that is “a media alternative for text and is clearly labeled as such,” meaning a video that presents nothing beyond text already on the page. Minutes summarize; the recording has every comment, every presentation and every exchange. The video needs its own captions.
Your meeting platform is still your content
Many districts stream through a meeting-video vendor or a board management portal, not YouTube. The rule covers content a district provides “directly or through contractual, licensing, or other arrangements,” so a vendor-hosted meeting page is yours. Ask your vendor whether they caption live and replays, who checks accuracy, whether speaker names are included, and whether they can take an uploaded caption file. If they can't, a YouTube copy with accurate captions linked from the meeting page is a practical fallback.
Old meetings already online
Years of past meetings are on most channels. The rules leave room here, but less than people assume:
- Embedded on your current board page (“Watch past meetings”): in scope unless you move them into a properly labeled archive that meets all four conditions of the archived content exception.
- Only on YouTube, uploaded before your deadline: probably exempt as preexisting social media posts, but unclear. DOJ didn't say whether YouTube counts.
- Any meeting someone asks for: must be made accessible, exempt or not. A resident researching a 2024 boundary vote is entitled to a captioned copy.
A practical approach: caption every meeting from now on, move older meetings to a clearly labeled archive page, and publish a request form for anything older.
How much work is it?
Correcting YouTube's auto-captions on a two-hour meeting, with speaker names, takes several hours. A district that holds two two-hour meetings a month has about 50 hours of meetings a year. That's why most districts either budget for a service or keep meetings shorter. If you do it yourself, start with how to add captions to YouTube videos. Ferndean's Standard plan ($2,400 a year, up to 60 hours) is built for districts that stream meetings: captions with board members named and motions written out, published back to YouTube, plus audio description where a meeting needs it. See pricing.
Sources
- DOJ final rule, 89 FR 31320 (April 24, 2024)
- WCAG 2.1, Understanding 1.2.4 Captions (Live)
- WCAG 2.1, Understanding 1.2.2 Captions (Prerecorded)
- WCAG 2.1, Understanding 1.2.5 Audio Description (Prerecorded)
- W3C WAI, Description of Visual Information
- YouTube Help, Use automatic captioning
This is a reading of the rules, not legal advice. Something out of date? Tell us and we'll fix it.