Whisper subtitle generation

No subtitles? No problem—Whisper listens to the audio and generates a subtitle track with AI.

Some YouTube videos don’t even have auto-captions. Whisper listens to the audio and uses AI to generate a subtitle track. This is a membership feature.

In the subtitle menu, this option is labeled Generate subtitles.

How it differs from AI Segmentation

  • AI Segmentation: Cleans up existing captions (auto, creator, or Netflix / Coursera official)—adds punctuation and re-breaks lines; doesn’t re-listen to audio. Free.
  • Whisper: When the video has no subtitles, generates them from audio. Members only.

In short: use AI Segmentation when captions exist; use Whisper when they don’t.

How to use

  1. Open a regular (non-live) YouTube video.
  2. Open the extension → Subtitles tab.
  3. In the subtitle menu, choose Generate subtitles.
  4. Wait for “Queuing” or “Generating” to finish. Subtitles appear—you can translate, practice, or export SRT as usual.

After generation, the system automatically applies AI Segmentation punctuation and line breaks, so Whisper subtitles are just as readable.

Support and limits

  • Supports 50+ languages—the set isn’t exactly the same as AI Segmentation (a few languages can’t be auto-segmented after generation).
  • Live streams not supported.
  • Very long videos may be rejected.
  • Membership required: Subscribe on Patreon and sign in in the extension.

Troubleshooting

  • Stuck in queue: Wait during peak hours, or try again later.
  • Generation failed: Confirm it’s not a live stream, the video isn’t too long, then retry.
  • Want to save time: If the video does have captions, AI Segmentation is usually faster.