Typing out subtitles by hand is one of the most tedious jobs in video editing. The good news is you almost never have to. CapCut auto captions on PC listen to your clip, transcribe the speech, and drop ready-to-edit text onto your timeline — usually in seconds. From there you tweak any mistakes, give the captions a look that fits your video, and export.

This guide to CapCut auto captions PC editing walks through the whole process: generating automatic subtitles, fixing the inevitable typos, styling the text so it reads well on phones, checking which languages are supported, and getting captions out either baked into the video or as a separate file. It works the same whether you're on Windows or Mac, as long as you're using the free desktop app from capcut.com.

Download CapCut for Desktop Get the latest official version from capcut.com
Quick answer
  • CapCut's auto captions feature transcribes your video's spoken audio into editable subtitles in a few seconds — no typing required.
  • Always proofread the generated text; auto transcription is fast but misses names, slang, and words spoken over background noise.
  • You can style captions with fonts, colors, and templates, then export the video with captions burned in or save them as a separate .srt file.

What auto captions do (and when to use them)

Auto captions take the spoken audio in your project and turn it into time-synced text. CapCut handles the listening, the transcription, and the timing — each line appears at roughly the moment the words are said, so you get a full set of subtitles without scrubbing the timeline and typing line by line.

It's the right tool whenever your video has clear speech: talking-head clips, tutorials, vlogs, interviews, voiceovers, and product walkthroughs all caption well. A huge share of social video gets watched on mute, so captions aren't just an accessibility nicety — they're often the difference between someone watching and someone scrolling past.

Auto captions transcribe speech, not music. If your clip is mostly a song or instrumental, the feature won't have much to work with. For lyric videos you'll want to type or paste the words manually instead.

Step 1: Add your video, then open Captions

Open CapCut and start a project the usual way — click New project, import your clip, and drag it onto the timeline. If you haven't installed the app yet, our step-by-step Windows 11 download guide gets you set up in a few minutes.

With your video on the timeline, look at the toolbar across the top of the editor and click the Captions tab (sometimes labeled Text with a captions section inside). You'll see an option along the lines of Auto captions or Auto-generate captions. That's the one you want.

Tip: If you only want captions for one part of the video, split the clip first and select just that segment before generating. CapCut transcribes the audio in your selection, so this keeps captions off any intro music or silent B-roll.

Step 2: Generate the automatic subtitles

When you open Auto captions, CapCut gives you a few quick choices before it starts:

  • Language: Pick the language actually spoken in the clip. Choosing the wrong one is the single most common reason captions come out as nonsense.
  • Source / audio track: If your project has more than one audio track — say, a voiceover and a music bed — tell CapCut which one carries the speech so it doesn't try to transcribe the song.
  • Sound-effect or filler options: Depending on your version, you may see toggles to clean up filler words or label sounds. These are optional.

Click Generate (or Create) and let it work. Short clips finish almost instantly; a long video takes a bit longer while CapCut transcribes the audio. When it's done, the captions appear as a connected text track on the timeline, already synced to the speech.

Some auto-caption features need an internet connection because part of the transcription happens online. If generation stalls or errors out, check that you're connected before trying again.

Step 3: Proofread and fix the text

Here's the step people skip and then regret. Automatic transcription is fast, but it is not perfect. It guesses at names, brands, slang, technical terms, and anything mumbled or spoken over background noise. Always read through before you publish.

CapCut makes editing painless. Open the caption list (often shown as a scrollable panel of all your subtitle lines) and you can:

  • Click any line to jump to it in the preview, so you can hear the audio and confirm the wording.
  • Edit text directly — fix a typo, correct a name, or rewrite an awkward auto-guess.
  • Merge or split lines so each caption is short enough to read comfortably.
  • Adjust timing by dragging a caption's edges on the timeline if a line appears too early or lingers too long.

A handy trick: use your editor's find-and-replace if a name or product is misspelled the same way throughout — fixing it once saves correcting it ten times.

Step 4: Style your captions so they actually read well

Default captions are functional but plain. A little styling makes them far easier to read, especially on a small phone screen. Select a caption (or select all of them to style every line at once) and open the text/style panel on the right. Useful controls include:

  • Font and size: Pick something clean and bold. Thin, decorative fonts are hard to read at a glance.
  • Color and outline: White text with a dark outline or a subtle drop shadow stays legible over almost any background.
  • Background bar: A semi-transparent box behind the words guarantees readability over busy footage.
  • Position: Keep captions in the lower third, but high enough that platform UI (like a TikTok caption or progress bar) won't cover them.
  • Caption templates / presets: CapCut includes preset styles and animated caption looks you can apply in one click, then fine-tune.

Set the style once and CapCut applies it across the whole caption set, so you're not formatting line by line. If you like working from ready-made looks, the same idea powers full project templates — see our guide to using CapCut templates on PC for more on that workflow.

Language support and accuracy tips

CapCut's auto captions support a broad list of languages — English, Spanish, Portuguese, French, German, Japanese, Korean, Indonesian, and many more — though the exact set and the accuracy can vary by language and update. The feature transcribes; it does not translate. If you want subtitles in a different language than the one spoken, you'd generate captions first and then translate the text (CapCut offers a translate option in many versions, or you can edit lines by hand).

Accuracy depends heavily on the quality of your audio. To get the cleanest transcription:

  • Use clear audio. A decent mic and minimal background noise make an enormous difference.
  • Pick the correct language before generating — and a single language per pass. Auto captions struggle when speakers switch languages mid-sentence.
  • Reduce music under speech. Loud backing tracks confuse the transcription. If you can, lower the music while generating, then bring it back up.
  • Speak at a steady pace. Very fast or overlapping speech is harder to capture cleanly.

No transcription engine is flawless. Treat auto captions as a strong first draft that saves you 90 percent of the typing — the last 10 percent is a quick proofread, and it's always worth it.

Step 5: Export with captions

Once your captions read well and look the way you want, you have two ways to get them out:

  1. Burned in (hardcoded): Just export the video normally. Because the captions sit on your timeline as text, they're rendered straight into the video file and show for everyone, on every player. This is the simplest option and the right choice for most social posts.
  2. As a separate subtitle file: CapCut can export captions as an .srt file alongside the video. Look for an Export captions or Export SRT option in or near the export dialog. A separate file lets viewers toggle subtitles on or off on platforms like YouTube, and it's easy to upload independently.

For the video itself, 1080p is a reliable default. When the export finishes, play it back and read the captions one more time on the final file — it's the last chance to catch a line that's mistimed or cut off.

Tip: If you sped a section up or slowed it down, re-check that captions still line up with the audio afterward. Our guide on changing video speed in CapCut covers how speed changes affect timing.

Common caption problems and quick fixes

A few issues come up again and again. Here's how to deal with them fast:

  • Captions came out as gibberish. Almost always the wrong language was selected. Delete the caption track and regenerate with the correct language.
  • It captioned the music instead of the voice. Point auto captions at the correct audio track, or lower the music before generating.
  • Lines are out of sync. Drag the edges of a caption block on the timeline to nudge it earlier or later. Big drifts usually trace back to a speed change made after captioning.
  • Generate button is greyed out or fails. Make sure a clip with audio is actually on the timeline and selected, and that you're online if your version needs a connection.
  • Captions get cut off at the screen edge. Move them up into the lower third and shorten any over-long lines so they don't wrap awkwardly.

Captions pair naturally with other polish steps. If you're also cleaning up your shots, our walkthrough on removing a video background in CapCut is a good next read.