How to Clip a Video in CapCut: A Beginner Workflow

Before you open CapCut

This is a general editing workflow, not a version-specific tour. CapCut’s mobile, desktop, and web interfaces differ, and feature availability can depend on your region, account, or plan. Check the current official interface for the exact label if a menu has moved.

Start with a file you own or are authorized to use. Keep the source unchanged and work on a project copy. Write down the start and end timestamps of your candidate moment so you do not spend the whole session searching the recording.

Import and isolate the moment

Create a new project and import the source video. Place it on the timeline. Move the playhead to the beginning of your selected moment and use the editor’s split control. Split again after the thought or action finishes.

Remove the unwanted material before and after the selection from the working timeline. Play the remaining cut. If it starts with an unexplained “that,” “he,” or “it,” restore enough context to identify the subject.

Refine the beginning before adding decorations. The first spoken phrase or visual should give a new viewer a reason to understand what follows. Trim repeated setup, but keep qualifications that affect meaning.

Set the canvas and follow the subject

Choose the aspect ratio for the destination. A 9:16 canvas is a common choice for vertical feeds. Resize and reposition the source image so the useful information fits the portrait frame.

Check the entire timeline. A centered crop that works for one speaker may hide the next speaker, a prop, or a gameplay event. Adjust position at the relevant points. Use keyframes only when a moving subject actually needs tracking; movement for its own sake can be distracting.

For a wide two-person conversation, consider alternating between speakers or using a layout that preserves both people. Do not crop so tightly that the body language explaining a reaction disappears.

Generate captions, then correct them

Find the caption or text controls and choose automatic captions for the spoken language. CapCut’s official documentation describes auto-caption support, but automatic transcription still needs review.

Play each caption against the voice. Correct names, quantities, contractions, and specialist vocabulary. Break long lines at natural phrases. Keep a consistent type size and avoid placing captions under platform buttons or the description area.

If there are multiple speakers, identify them in a way that does not rely on color alone. A small speaker name or stable placement can supplement the color scheme. Read more in the caption guide.

Balance sound before adding effects

Listen for uneven dialogue, background hum, and music that competes with the voice. Start with a comfortable, consistent speech level. If a noise-reduction control is available, compare the processed and original audio; aggressive processing can make speech sound unnatural.

Only add sound effects when they help communicate the moment. Sudden volume spikes can make an otherwise good clip unpleasant to watch. Check with phone speakers, because an edit that seems clear in headphones can still sound muddy on a phone.

Export and inspect

Choose resolution and frame rate appropriate to your footage and destination. For a 1080p portrait workflow, 1080 × 1920 is a reasonable starting point when the source supports it. Upscaling low-resolution footage does not restore lost detail.

Watch the exported file from beginning to end. Look for caption timing changes, accidental black frames, missing audio, and a cut that ends before the final word. Also inspect whether the chosen features introduced a watermark or require a plan upgrade before publishing.

Use the clip checklist and keep your project file. Save a reusable project only after the first export works. A template is useful when it preserves good defaults, not when it locks every clip into the same inappropriate crop or pacing.

If auto captions do not work

Check whether text failed to generate or whether generated text is hidden. The caption troubleshooting guide provides a small-sample test and separate checks for generation, visibility, and export.