CaptaHow to add captions to a screen recording on a Mac
Five steps to captioned narration, transcribed on your Mac and burned into the export.
Most people watch product videos with the sound off, so a screen recording without captions loses its narration entirely. The usual route is a cloud transcription service, which means uploading the recording. This guide adds captions with Capta, a free macOS screen recorder that transcribes narration on the Mac itself and burns the captions into the export.
Captions are transcribed on your Mac by a model you download once (1.84 GB; the Voice tab offers it the moment you switch the mic on), or by Apple Speech on macOS 26 or later. If transcription fails, use Voice → Transcribe again after the speech model is installed. A failed retry keeps existing text; a successful retry backs up the previous captions.
What you need
- A Mac with macOS 14 or later and Apple silicon.
- Capta, free, 52.4 MB.
- A microphone, and the one-time caption download the Voice tab offers when the mic is on (or Apple Speech on macOS 26 or later). Transcription runs on the Mac afterwards.
Switch the microphone on before you record
Open the capture panel with ⌥⌘R and choose a microphone under Options. The built-in microphone is fine for a demo; if you have an external one, pick it there. The mic is off by default. The preflight meter lets you check levels; audio is only saved once recording starts.
Talk while you record
Press ⌥⌘R and narrate in plain sentences: what you are clicking and why it matters. Speak a little slower than usual and finish each sentence. Capta steers its cuts away from your sentences, so complete ones give it more to work with. Stop with ⌥⌘R.
Read the captions on the words lane
After you stop, Capta transcribes with your selected Qwen or Apple Speech model. Wait for transcription to finish; the audio stays on your Mac. Each line appears on the words lane of the timeline, timed to when you said it, and the same lines are listed in the Voice tab. The music, if you added one, is lowered under your voice automatically.
Fix any word
Speech models get product names and jargon wrong more often than ordinary words. Click a line in the Voice tab, retype it, and the burned-in caption updates in the preview. Long transcripts scroll inside the tab.
Export with captions burned in
Press Export. Captions are rendered into the frames, so they show in every player and in social feeds that autoplay muted, which is where most people will meet your video. For a square or vertical social cut, pick that size in Export; the captions re-flow to fit.
Common questions
Qwen word alignment supports English, Chinese, Cantonese, French, German, Italian, Portuguese, Russian and Spanish. Other supported transcription languages, including Greek, use approximate timing. Review captions before sharing.
Do I need Apple Intelligence for captions?
No. Captions come from the downloaded caption model or Apple Speech, independently of the writing model. Titles, the post and chapter names come from built-in templates, from a small model you download once, or from Apple Intelligence; you can rename any chapter and move the key moment by hand in the Edit tab.
Can I change how the captions look?
Yes. Voice → Caption look: a pill, a bar across the frame or plain text, at the bottom or the top, with a size slider. The subtitle file you export is unaffected.
Can I add captions to a recording I already made?
Captions come from the narration Capta recorded, so a take made with the mic off has no words to transcribe. Record the take again with Mic on, or narrate a fresh take over the same flow.
Where does the transcript go?
Into the recording's own folder in ~/Movies/Capta, next to the video, and nowhere else. The privacy page has the details.
One shortcut to record, one click to export.