How to caption lecture videos on a Mac
To caption a lecture video on a Mac: get an automatic transcript, correct it, save it as a caption file (.srt or .vtt), and upload that file next to the video in your course platform. The correcting step is the one people skip, and it's the one that matters. If you teach at a public college or university in the US, there's also a deadline now. Here's the whole process, including how to make an SRT file by hand and why Canvas wants VTT for iPhones.
Why captions are on your desk now
Under the US Department of Justice's ADA Title II rule, public entities, including public colleges and universities, have to make web content and course materials meet WCAG 2.1 Level AA. For recorded video, that standard includes captions (success criterion 1.2.2). In April 2026 the DOJ pushed the compliance dates back by a year:
| Who | Compliance date |
|---|---|
| Public entities serving 50,000 people or more | April 26, 2027 |
| Smaller public entities and special districts | April 26, 2028 |
The extension moved the dates, not the requirement. This isn't legal advice, and the details (including which archived content is exempt) matter, so talk to your institution's accessibility office. They often have captioning tools or budget you can use.
Four ways to get captions on a Mac
| Option | Cost to you | Where the audio goes |
|---|---|---|
| Your school's video platform (Panopto, Kaltura, Canvas Studio) | Usually free | Your school's platform |
| YouTube: upload as unlisted, edit the auto-captions, download the .srt | Free | |
| A Mac transcription app, such as MacWhisper | Free or paid tiers | Stays on your Mac |
| A screen recorder that captions what it records | Depends on the app | Depends on the app |
Start with what your school already pays for. If your lectures already live in Panopto or Kaltura, use their automatic captions and built-in caption editor. That's the least work. The other options matter when you record outside the school's system or don't want to upload a recording before it's ready.
Automatic captions are a draft
Automatic speech recognition is good at everyday speech and bad at exactly what lectures are full of: names, technical terms, acronyms, numbers and formulas. "Acetyl-CoA" can easily come out as "a seat of co-A." A caption with a wrong term isn't just untidy. It teaches the wrong word to the student relying on it.
- Review at 1.25× speed with the text in front of you. Most errors cluster around the same terms.
- Keep a course glossary. Once you've fixed "Krebs cycle" ten times, you'll know to search for it.
- Check numbers and units. "Fifteen" and "fifty" sound alike. A formula read aloud rarely transcribes cleanly.
- Mark who's speaking when it changes, especially in recorded discussions.
What an SRT file is, and how to make one by hand
An SRT file is plain text: a number, a time range, and the caption text, separated by a blank line.
1
00:00:00,500 --> 00:00:03,000
Today we cover the Krebs cycle.
2
00:00:03,200 --> 00:00:06,000
First, acetyl-CoA enters the cycle.
To write or fix one on a Mac with no extra software:
- Open TextEdit and choose Format → Make Plain Text. Rich text breaks caption files.
- Type or paste the captions in the format above. Timestamps use a comma before the milliseconds.
- Save, and name the file with
.srtat the end. If TextEdit offers to add.txt, choose to keep.srt.
Typing a whole lecture this way is slow. In practice you'll do it to fix a file a tool already produced, not to write one from scratch.
Uploading to Canvas, and why you want VTT too
Canvas accepts caption files in both SRT and WebVTT (.vtt) format (Stanford's Canvas captioning guide covers the upload steps). WebVTT is the format Apple devices use, and Rev's Canvas guide points to WebVTT for iOS. Several university help pages warn that SRT captions may not show in the Canvas iPhone app. Uploading a .vtt avoids the question.
Converting takes one command with ffmpeg (brew install ffmpeg). I tested it on the example above:
ffmpeg -i lecture.srt lecture.vtt
Panopto, Kaltura and YouTube all accept SRT directly.
Record lectures that caption well
Better audio means fewer corrections later.
- Mic close to your mouth. A laptop's built-in mic across the room picks up echo, and that's what transcription gets wrong.
- Repeat student questions before answering. The mic won't pick them up, and the captions shouldn't skip half the conversation.
- Put key terms on screen as you say them. It helps everyone, and you can check spellings against your own slides.
- Record in short segments. A 10-minute file is much faster to review than a 75-minute lecture.
The tools, honestly
If your school gives you Panopto or Kaltura with captioning, use it. It's free to you and the files already live where students watch them.
I make CursorFlow, a Mac screen recorder, and it's useful when you record screen-based lessons on your own Mac:
- It transcribes as part of the recording workflow, using Apple's Speech framework. For languages your Mac supports on-device, the audio never leaves it. For other languages, Apple's speech service does the transcription.
- You can edit caption text and timing in the editor, which is where the review step happens.
- It saves a .srt file next to the exported video, or burns the captions into the picture.
- It doesn't export .vtt. Use the ffmpeg command above for Canvas.
It's free to use, with a watermark on exports until you buy the one-time licence. More detail on how Mac transcription works, and where it falls short, is in this post on on-device captions. For full courses, see how to record an online course on a Mac.


