Skip to content
CapCut logo

CapCut

Video editing with AI captions, effects, and generative clips.

Video

How it works / How to use

Video editing with AI captions, effects, and generative clips. Import footage on CapCut Web, Desktop, or Mobile, edit on the timeline, add AI captions or effects where available, then export or share. Features and Pro benefits vary by platform and plan—check CapCut help for your device.

  1. Open CapCut on web, desktop, or mobile and create or open a project.
  2. Import video, audio, and images onto the timeline.
  3. Edit clips—trim, arrange, add text, captions, transitions, and AI tools supported on your platform.
  4. Preview the full timeline, set export range if needed, then export or share.

General video editing workflow

Import media, build the story on the timeline, layer text and audio, preview end-to-end, then export with the resolution and format your platform needs.

What to provide

  • Source clips, audio, and target platform (Reels, TikTok, YouTube)
  • Desired length and pacing
  • Brand elements: fonts, colors, logo usage

Details that improve the result

  • CapCut runs on web, Windows, macOS, iOS, and Android with platform-specific UI
  • Use In and Out markers on desktop to export only part of the timeline
  • CapCut does not import or export Adobe Premiere Pro project files per official help

Example prompt

30-second product teaser: three b-roll clips, logo end card, background music under dialogue, 1080p export for Instagram Reels.

If the first output is not good

  • “Trim clip 2 by 0.5s; keep music sync.”
  • “Move text lower third on all caption blocks.”
  • “Export only the marked In/Out section.”

Common mistakes

  • Editing without checking final aspect ratio for the target platform
  • Stacking many effects before previewing playback performance
  • Assuming desktop template search works like mobile or web

Auto captions and subtitles

Import video, open Captions > Auto Captions (or Text > Auto captions on some surfaces), choose language, generate, then review and edit text and timing on the timeline.

What to provide

  • Video with clear spoken audio
  • Spoken language (and bilingual option if available)
  • Caption style: font, size, animation, placement

Details that improve the result

  • Auto Caption (Recognise Subtitles) uses AI speech-to-text on web, desktop, and mobile
  • Always review auto-generated text; fix timing and spelling before export
  • Advanced options like speaker detection or batch SRT export may vary by platform and region

Example prompt

Podcast clip with English speech: generate auto captions, fix names in the first 10 seconds, use bold keyword highlight if available, export with embedded captions.

If the first output is not good

  • “Split long caption lines at natural pauses.”
  • “Match caption font to brand: Montserrat, white, dark outline.”
  • “Export sidecar SRT if supported on desktop for YouTube upload.”

Common mistakes

  • Publishing without proofreading auto captions
  • Generating captions before cleaning noisy background audio
  • Expecting identical caption features on every platform version

AI voiceover and ad scripts

CapCut’s help articles describe adding AI voiceover and captions to ad videos: write and verify the script, generate voice, auto-caption, proofread, then export.

What to provide

  • Script text with exact wording and pronunciation notes
  • Voice tone and pacing preferences supported in the tool
  • Legal or brand claims that must appear verbatim

Details that improve the result

  • Verify script compliance before generating voiceover
  • Pair voiceover with auto captions for accessibility and muted playback
  • Review timing between voice and on-screen text before export

Example prompt

30-second app install ad: paste approved script, generate AI voiceover in English, run auto captions from the same audio, proofread claims and disclaimers, export 1080x1920.

If the first output is not good

  • “Slow voiceover slightly; regenerate captions after.”
  • “Fix product name capitalization in all caption blocks.”
  • “Shorten script by 15 words; keep CTA verbatim.”

Common mistakes

  • Letting AI rewrite regulated ad copy without review
  • Skipping caption pass after changing voiceover timing
  • Mixing multiple languages without setting bilingual options correctly

Auto Cut beat-sync editing

Auto Cut is an AI feature that trims and syncs clips to music beats, speech pauses, or text prompts. Generate a first cut, then manually adjust on the timeline.

What to provide

  • Raw clips and a music track—or a text prompt describing pacing
  • Target vibe: fast social cut vs slower narrative
  • Clips you want prioritized in the edit

Details that improve the result

  • Official help describes Auto Cut as AI-powered trimming and sync
  • Treat the first Auto Cut pass as a draft, not a final edit
  • Check transitions and clip order after automation

Example prompt

Travel montage: import 12 clips and background track, run Auto Cut synced to beats, remove one repeated clip, tighten opening two seconds manually.

If the first output is not good

  • “Re-run Auto Cut with slower pacing prompt.”
  • “Keep clip 5 and 8 fixed; re-sync the rest.”
  • “Add manual transition only between sections.”

Common mistakes

  • Exporting Auto Cut output without watching full preview
  • Using mismatched music tempo and clip lengths
  • Expecting Auto Cut to replace story decisions entirely

Templates (mobile and web)

Browse or search templates on mobile or web, open a template, replace media and text slots, preview timing, then export. Profile templates are creator designs, not finished videos.

What to provide

  • Template platform: CapCut mobile or web (desktop has limited built-in templates without search)
  • Footage or photos to replace placeholder slots
  • Text and branding to swap in

Details that improve the result

  • Template search is on mobile and web; desktop shows a limited built-in set without search per help center
  • Some downloaded templates on desktop may show a single frame until slots are filled
  • Remove default ending clips if unwanted watermarks appear

Example prompt

On CapCut mobile: search trending TikTok template, replace three photo slots, update headline text, preview at full speed, export without default watermark ending if present.

If the first output is not good

  • “Swap template music; keep beat markers.”
  • “Replace only text layers; keep motion unchanged.”
  • “Duplicate template project for A/B headline tests.”

Common mistakes

  • Searching templates on desktop expecting full catalog search
  • Leaving placeholder media in export
  • Ignoring template license or watermark elements at the end

Export partial clips with In/Out markers

On CapCut desktop, set In and Out markers to define the range, then export only that section instead of the full timeline.

What to provide

  • Timeline section to export
  • In and Out points on desktop
  • Export resolution, format, and quality settings

Details that improve the result

  • Useful for social clips cut from long recordings
  • Confirm markers include full intro/outro handles you need
  • Check caption and audio tracks span the marked range

Example prompt

From a 20-minute interview timeline, mark In at 03:12 and Out at 04:05 for a teaser, export 1080p with embedded captions.

If the first output is not good

  • “Extend Out by 0.3s to finish the sentence.”
  • “Export same range as SRT sidecar if available.”
  • “Duplicate markers for alternate 15s cut.”

Common mistakes

  • Exporting full timeline when only a segment is needed
  • Cutting mid-word because Out marker is tight
  • Forgetting nested text or effects outside the marked range

AI image and design generation

Use CapCut’s AI design or image tools where available on your platform. Expect iteration: CapCut help notes generated text and subjects can vary between runs.

What to provide

  • Clear visual prompt for the asset you need
  • Aspect ratio or placement in the video
  • Text to appear on screen, spelled exactly—or avoid text in prompts when unreliable

Details that improve the result

  • Garbled text in AI images is a known limitation when simulating lettering
  • Regenerate or edit manually when people or objects shift unexpectedly
  • Combine AI visuals with manual layout for polished results

Example prompt

Generate a simple background illustration for a title card: minimal flat shapes, brand blue palette, no text in the image. Composite headline text manually in CapCut.

If the first output is not good

  • “Same palette; simpler composition with fewer objects.”
  • “Regenerate without people; keep abstract shapes only.”
  • “Add manual text layer instead of text inside AI image.”

Common mistakes

  • Expecting perfect on-image typography from AI generation
  • Using AI output without checking consistency across scenes
  • Replacing entire edit when one regenerate would suffice

Background removal and cleanup tools

CapCut help documents making a photo background white and similar cleanup flows: remove background, add solid color or new layer, refine edges, export still or composite on timeline.

What to provide

  • Photo or video clip needing background change
  • Target background color or replacement asset
  • Edge quality requirements (hair, glass, motion)

Details that improve the result

  • Review edges after automatic removal
  • For video, check flicker frame-by-frame on fine detail
  • Combine with captions or overlays only after base visual is clean

Example prompt

Product photo on timeline: remove original background, add solid white, soften edges on the mug handle, place on title slide before export.

If the first output is not good

  • “Feather edges slightly; keep product color unchanged.”
  • “Replace white with soft gradient background.”
  • “Fix one frame where mask slips on motion clip.”

Common mistakes

  • Accepting rough masks on hair or transparent objects
  • Changing background before fixing exposure on subject
  • Applying heavy effects before isolation is clean

How to prompt

CapCut AI features respond to clear creative briefs: subject, pacing, language, style, and what to exclude. For captions and voiceover, the script is the prompt—spell names, claims, and CTAs exactly and proofread after generation.

Task: 25s Instagram Reel.
Footage: 4 gym clips, upbeat track.
Captions: English auto captions, bold keywords, bottom safe zone.
Exclude: default template watermark ending.
Export: 1080x1920, embedded captions.

Write scripts verbatim for voice and captions

AI voiceover and captions follow your script text. Put pronunciation, product names, and legal lines in the document you generate from.

Example

Script: “Try FitTrack free for 14 days. Cancel anytime.” Generate voiceover, then auto captions from the same audio and fix only timing—not marketing claims.

Describe pacing for Auto Cut

Say whether you want a fast beat-sync montage or slower narrative rhythm so Auto Cut has a clear target.

Example

Auto Cut: sync 8 travel clips to chorus drops, medium-fast pacing, avoid repeating the same location back-to-back.

Specify language before caption generation

Choose spoken language (and bilingual options when available) before generating so recognition starts accurate.

Example

Spoken language: Spanish (Mexico). Generate auto captions, then translate to English subtitles if bilingual export is enabled.

Keep AI visuals text-free when possible

CapCut help notes AI-generated text can garble. Prompt for clean visuals and add readable text as editable layers.

Example

AI background: abstract gradient waves, navy and teal, no letters, no logos. Add title with CapCut text tool.

State platform and export target

Aspect ratio, duration, and embedded vs sidecar captions depend on where the video will publish.

Example

YouTube long-form segment: mark In/Out 02:10–05:40, export 1080p with SRT sidecar if desktop supports it for upload.

Best output tips

Know your platform differences

CapCut Web, Desktop, and Mobile share core editing but differ in template search, caption extras, and some AI tools. Read help articles for your device before starting.

Build on the timeline first

Arrange story, audio, and pacing before stacking AI features. A clean timeline makes caption and export passes faster.

Run auto captions on clean audio

Background noise and overlapping speakers reduce recognition quality. Use the clearest dialogue track you have before Generate.

Always proofread captions

Auto Caption transcribes speech automatically but can miss names, homophones, and punctuation. Review timing and text before publishing.

Style captions for readability

Adjust font, size, contrast, and safe margins for mobile playback. Use keyword highlight or animation sparingly for clarity.

Use bilingual options when needed

Desktop help documents bilingual subtitles and keyword highlight on auto captions. Set languages before generation when exporting multilingual cuts.

Draft with Auto Cut

Auto Cut syncs clips to beats or prompts. Watch the full preview and manually fix order, duration, and transitions afterward.

Export with purpose

Set resolution, format, and quality for the destination platform. Export embedded captions or sidecar SRT/VTT when supported for accessibility.

Clip exports with In/Out

On desktop, mark In and Out to export a segment from a longer timeline without duplicating the project.

Templates on mobile and web

Search trending templates on mobile or web, replace slots, and remove default ending clips if watermarks appear. Desktop built-in templates are more limited.

Scripts drive AI voiceover

Write and verify ad or narration scripts before generation. Regenerate voice after script changes, then refresh captions to match.

Avoid AI text inside images

Help articles note garbled glyphs when AI simulates text. Generate clean visuals and set readable text with CapCut’s text tools.

Iterate AI visuals

People, objects, and colors can shift between generations. Regenerate or adjust prompts slightly instead of over-editing a poor first result.

Understand subscription scope

CapCut Pro and Standard benefits differ by platform and region. Official help explains cross-platform access and diamond-icon items—verify before purchase.

No Premiere project exchange

CapCut does not support importing or exporting Adobe Premiere Pro projects. Plan manual media export and rebuild if switching tools.

  • Import and organize media before adding captions or heavy effects.
  • Generate auto captions, then proofread every line—especially names and numbers.
  • Pick spoken language before caption generation for better accuracy.
  • Use In/Out markers on desktop to export clips without re-editing the master.
  • Treat Auto Cut and AI edits as first drafts; refine on the timeline.
  • Search templates on mobile or web; desktop template search is limited per help docs.
  • Add on-screen text manually when AI image text renders poorly.
  • Check Pro vs Standard plan features for your platform before relying on cross-device benefits.

Try this AI

Try CapCut

Product Details

Pricing, features, limits and latest updates

CapCut

Video editing with AI captions, effects, and generative clips.

Free / Paid · Free

Pricing Plans

Free

Free

Pro

$19.99/ Monthly

Pro

$179.99/ Yearly

Key Features

Video Generation

AI video, captions, and effects sit inside a timeline editor; some actions spend AI credits.

Automation

Auto Captions and Auto Cut are core AI editing workflows; platform availability differs between Web, Desktop, and Mobile.

Image Editing

AI image tools and cutout features are available in the editor and may be credit-gated.

Limits

  • Auto Cut platforms: Auto Cut is documented on mobile and desktop; it is not supported on CapCut Web (CapCut Help Center).
  • Pro simultaneous devices: 5 devices. CapCut Pro can be used on up to five devices simultaneously when logged into the same account (CapCut Help Center).

Ideas / Prompt experiences

Share a prompt that worked for you. Username and email are shown with your submission. External links are not allowed.

Example prompt

Auto Captions on a talking-head clip

Prompt

Workflow: Import a 60-second interview clip → Auto Captions → review transcript → fix names and numbers manually → export 16:9 for social.

Short explanation

CapCut help recommends proofreading AI captions; speaker names and numbers often need manual correction.

Example prompt

Auto Cut beat-sync (mobile/desktop)

Prompt

Workflow: Templates → AI Creation → Auto Cut → select 8–12 B-roll clips → choose music-driven sync → review cuts before export.

Short explanation

Auto Cut is documented on mobile and desktop only, not CapCut Web as of official help articles.

Share your experience

Required fields are marked. Variation, result, and explanation are optional.

Pika logo

Pika

Explore

AI video platform for text-to-video and image-to-video generation with credit-based plans and tools including Pikascenes, Pikadditions, Pikaswaps, and Pikatwists.

VideoFree / Paid

Synthesia logo

Synthesia

Explore

AI presenter videos from a script, built for training and explainers.

VideoFree / Paid

Adobe Podcast logo

AI audio recording and editing on the web, including Enhance Speech for noise removal and Studio for browser-based podcast production.

Video