How to make subtitles from a video with AI on a Mac

Turn the speech in any video into timed subtitles with OpenAI's Whisper model, then clean them up and export SRT — on a Mac, without a subscription.

Updated 2026-10-07 · Code Boop

Typing subtitles by hand takes hours per film. Speech recognition does it in minutes: OpenAI's Whisper model writes down what is said, with a start and end time for every line. The result is a subtitle file you can edit, translate and burn in.

Step by step

  1. Open the video in AI Caption & Translate Pro (MP4, MOV, MKV and most other formats).
  2. Click Transcribe. Choose the service — Groq's free tier or OpenAI, with your own key — and the spoken language, or let it be detected.
  3. Wait a few minutes. The audio is sent in parts to the service you chose and comes back as timed subtitle lines. The new file is saved as “Movie.en.srt” in a “Movie Subtitles” folder next to the video.
  4. Clean it up. Clean Up Transcription (⇧⌘K) removes repeated and empty lines and recognition artefacts, fixes spelling, and can correct misheard words with AI. You see every change before it is applied.
  5. Fine-tune the timing on the audio waveform: drag the edges of a line, or use the nudge buttons.

Accuracy tips

  • Choose the spoken language instead of auto-detect when the film starts with music or silence.
  • Noisy sound? Clean it first with AI Sound Restore Pro — clearer dialogue means fewer recognition errors.
  • Only the audio is sent for transcription; the video itself stays on your Mac.

Next step: translate the subtitles into other languages.

More guides