• Success Criterion 1.2.2
  • Conformance level A

Multi-Speaker Video Captions With No Speaker Labels

1.2.2 — Captions (Prerecorded)

Scenario

Setting

An e-book reader's library page

What’s wrong

Captions missing speaker identification in multi-speaker content.

Example 

00:00:10.000 --> 00:00:13.000
So what made you want to write this book?
I wanted to explore grief through a fantasy lens.

REM Two different speakers (interviewer and author) merged into one
REM caption block with no labels.

Why it matters

A deaf reader watching the author interview can't tell where the interviewer's question ends and the author's answer begins, so the exchange reads as one confusing run-on.

How to test

Play any video with multiple speakers: if captions don't indicate who is speaking (name or label), it fails.

How to fix

Multi-speaker captions need a clear speaker label on every line change, even a simple all-caps name works.

00:00:10.000 --> 00:00:11.500
INTERVIEWER: So what made you want to write this book?

00:00:12.000 --> 00:00:14.000
AUTHOR: I wanted to explore grief through a fantasy lens.

Outcome

A deaf reader follows the interview exchange and understands exactly which answer belongs to the author.

Who is affected

Deaf and hard-of-hearing viewers following captions lose track of who is speaking whenever more than one voice appears.

Learn more

Related scenarios