Key takeaways
- CapCut can create automatic captions on mobile, desktop and the web, although menu names can vary slightly.
- Clean speech, the correct language and one active voice track produce more accurate captions.
- Edit the transcript before changing fonts, colors or animation, so corrections take less time.
- Always watch the exported file with sound off to catch missing captions, poor timing and unreadable text.
Quick answer
To learn how to add auto captions in CapCut, import your video, open the Text or Captions panel, choose Auto captions, select the spoken language, then generate and review the text. Correct names, timing and line breaks before styling the captions and exporting the finished video.
How to add auto captions in CapCut on desktop, web and mobile
CapCut turns speech in a video into timed text. The basic process is similar on every device, but the buttons can sit in different places after an app update.
- Start a project. Open CapCut and create a new project. Import your video, then place it on the timeline.
- Check the audio. Play a short section and confirm that the spoken track is audible. Mute duplicate voice tracks, background videos and any music that covers the speaker.
- Open captions. On desktop or the web, choose Captions or Text, then Auto captions. On mobile, tap Text or Captions in the bottom toolbar, then tap Auto captions.
- Choose the source. Select the original video sound, a voiceover or both, depending on where the speech sits. Pick both only when each source contains words that viewers need to read.
- Select the language. Choose the language actually spoken in the recording. Do not choose a translation language at this stage unless you want translated subtitles.
- Generate the captions. Tap or click Generate. CapCut analyses the audio and places separate caption clips above the video track.
- Review every line. Play from the beginning and correct spelling, punctuation, speaker names and timing. Split long captions and join fragments where needed.
- Style and export. Apply one readable style across all caption clips, preview the full video, then export it. Check the saved file rather than relying only on the editing preview.
For a long video, generate captions after the main edit is mostly finished. If you cut or move many scenes later, inspect the caption timing again because some clips may no longer match the speech.
What should you do before generating captions?
Automatic captioning depends on the audio it receives. A clear voice recorded close to the microphone usually needs fewer corrections than speech mixed with loud music, echo or traffic.
Prepare the timeline
- Finish major cuts first. Remove failed takes, long pauses and sections you will not publish.
- Reduce competing sound. Lower music while someone speaks. If CapCut offers voice enhancement or noise reduction in your version, test it on a short section before applying it to the full project.
- Choose one useful speech source. If the camera audio and an external microphone contain the same words, mute the weaker track. Two copies of one voice can create echoes and repeated text.
- Check the language. Pick the main spoken language. Mixed language videos often need manual corrections, especially for names, slang and technical terms.
- Keep the app connected. Auto caption processing may require an internet connection. Avoid starting a large upload on a weak mobile connection.
Listen through headphones before generating. If you cannot easily understand a sentence, the transcription system may struggle with it too.
Plan for short and long videos
For a short vertical video, captions should be large enough for a phone screen and should not cover the speaker's mouth, product or platform buttons. Keep each caption to one or two short lines when possible.
For a long tutorial, use restrained styling and comfortable line lengths. Generate captions by chapter if a very large project causes slow processing, then keep the same text style across every section.
How do you correct and style CapCut captions?
Correct the words before styling them. This prevents you from spending time formatting caption clips that may later be split, joined or deleted.
Edit the transcript
Select a caption clip on the timeline or open the caption list. Replace incorrect words, add useful punctuation and fix capital letters, but keep the text close to what the speaker actually says.
Pay special attention to names, prices, product terms and calls to action. Auto captions may turn a web address, brand name or unfamiliar phrase into common words that sound similar.
Break long text at natural pauses. A useful caption shows a complete thought without forcing the viewer to read a full paragraph in a second or two.
Adjust timing
Drag the beginning or end of a caption clip so it appears when the phrase is spoken. Avoid leaving an old caption on screen after the next sentence begins.
If one clip contains two distinct sentences, use the split control and edit each part. If CapCut creates several one word fragments, combine or extend them so the captions do not flash too quickly.
Apply a readable style
- Font. Use a simple, thick font that remains readable on a small screen.
- Size. Preview at normal phone size, not only in a large editor window.
- Contrast. Use light text with a dark outline or background, or dark text on a light box.
- Position. Keep text inside the safe area and away from faces, lower screen controls and important visuals.
- Animation. Use subtle motion. Busy word effects can distract from a tutorial or long interview.
CapCut may let you apply a style to all caption clips. Confirm that the change affects the full set, because selecting one clip can sometimes edit only that caption.
Why are CapCut auto captions wrong, missing or out of sync?
Most caption problems come from the audio source, language setting, timeline edits or processing connection. Start with the simplest check before deleting the full transcript.
| Problem | Likely cause | What to try |
|---|---|---|
| No captions appear | The wrong audio source was selected, the track is muted or processing failed. | Play the voice track, select its source, check the connection and generate again. |
| Words are inaccurate | Music, echo, accents, mixed languages or unusual names are confusing the transcription. | Lower background sound, choose the correct language and edit names manually. |
| Text is duplicated | Two audio tracks contain the same speech, or captions were generated twice. | Mute the duplicate audio and delete the extra caption set. |
| Captions drift out of sync | The video was trimmed, sped up or moved after captions were created. | Regenerate after the final edit, or retime the affected caption clips. |
| Captions cut off words | A clip boundary is too short or the text box is too narrow. | Extend the clip, split at a natural pause or reduce the font size slightly. |
| Auto captions gets stuck | The connection, device storage or project size is causing trouble. | Save the project, restart CapCut, free storage and process shorter sections. |
| Captions disappear after export | The caption track is hidden, outside the frame or not included in the chosen export method. | Make the track visible, move text inside the canvas and test a short export. |
If one section has poor sound, correct that section manually instead of regenerating an otherwise accurate transcript. Regenerating can replace edits you already made, depending on the CapCut version and the option selected.
For an app problem, save the project before clearing a cache, updating CapCut or reinstalling it. A local draft may not be backed up unless you deliberately saved it to your account or cloud space.
How should you export captions, and what does CapCut cost?
Most social videos use burned in captions, which means the words become part of the picture. Viewers cannot turn them off, but they will appear on platforms that do not preserve a separate subtitle file.
A subtitle file, usually SRT, stores the words and time codes separately. This is useful for YouTube, client review and translation, but caption file export can depend on the device, region, account and CapCut plan.
Choose the right output
| Output | Best for | Check before delivery |
|---|---|---|
| Video with burned in captions | TikTok, Instagram Reels, YouTube Shorts and social ads | Text position, contrast, spelling and export resolution |
| SRT subtitle file | YouTube, courses, client handoff and later translation | Time codes, line breaks, file encoding and speaker names |
| Video plus SRT | Archives and flexible client delivery | The video and subtitle file use matching names and timing |
Look for Export captions or a download option in the caption panel. If it is missing on mobile, open the project in CapCut desktop or web and check there, as available controls can differ.
Cost and alternatives
CapCut offers a free plan, and some auto caption features may be available without payment. CapCut Pro pricing varies by platform and region, but it is often around $20 per month at the time of writing, with annual billing sometimes costing less per month.
Check the checkout screen before subscribing, since AI usage limits, trials and included features can change. Do not pay only to create one transcript until you have confirmed that your account requires Pro for the specific caption style, translation or export option you need.
If CapCut cannot process a project, YouTube Studio can generate captions after upload, while Descript and Adobe Premiere Pro offer transcript based editing and caption tools. YouTube Studio is useful for platform subtitles, Descript commonly starts around $20 per month, and Adobe Premiere Pro commonly starts around $23 per month, with prices varying at the time of writing.
FAQ
Are CapCut auto captions free?
CapCut often includes basic auto caption tools on its free plan, but availability can vary by device, region and account. Some styles, AI features, translation options or exports may require CapCut Pro, so check for a Pro label before processing a full video.
Can CapCut add captions in another language?
CapCut may offer caption translation for supported languages, depending on the version and plan. First create and correct captions in the spoken language, then translate them so names and important terms are not mistranslated at the first step.
Can I export CapCut captions as an SRT file?
CapCut desktop or web may provide an option to export generated captions as a subtitle file. If the option is not visible, check that the captions were created as editable auto captions rather than added as ordinary text, and confirm whether your plan supports the export.
Why does CapCut say there is no speech detected?
The selected source may be silent, muted or covered by other sound. Confirm that the voice plays in the timeline, choose the correct original sound or voiceover source, then try a short section with a stable internet connection.
Will captions stay synced if I edit the video later?
Small text corrections will not usually disturb timing, but trimming, moving or changing the speed of video clips can put captions out of sync. Complete major video edits first, then generate captions or inspect every affected section afterward.