How to Add Auto Captions to a Video in CapCut

Use CapCut's Auto Captions feature to transcribe spoken audio into editable, styled subtitles in a couple of minutes.

6 minBeginner

Visit CapCut

Auto Captions listens to the speech in your timeline and turns it into a synced subtitle track you can edit and restyle. It is the single highest-impact thing you can do for short-form video, since most people watch with the sound off. This guide covers the desktop app, but the steps are nearly identical in the mobile app.

What you need

The CapCut desktop or mobile app, signed in to a free account
A clip with clear spoken audio already on the timeline
An internet connection (the transcription runs in the cloud)

Step 1: Drop your clip on the timeline

Create a new project, then drag your video from the media pool down to the timeline. Auto Captions reads whatever audio is on the timeline, so make sure the clip with the talking is actually placed, not just imported.

CapCut - Project workspace

[ Media ] [ Audio ] [ Text ] [ Stickers ] | Player

------------------------------------------------+----------

Timeline | [ video ]

V1 |#### interview_take2.mp4 ################| |

A1 |~~~~~~~~ waveform ~~~~~~~~~~~~~~~~~~~~~~~| |

The clip sits on the video track with its waveform visible.

Step 2: Open the Captions panel

In the top toolbar click Captions, then choose Auto captions. A small dialog appears where you pick the spoken language and which audio source to transcribe.

Step 3: Pick language and generate

Set the spoken language to match your audio. If your project mixes a voiceover and original sound, choose the source that actually contains the speech. Click Generate and wait a few seconds while CapCut uploads, transcribes, and drops a text track onto the timeline.

CapCut - Auto captions dialog

Auto captions

----------------------------------------

Spoken language: [ English (US) v ]

Source: [ Original sound v ]

[x] Clear existing captions first

[ Cancel ] [ Generate ]

Step 4: Fix mistakes and restyle

Double-click any caption to correct misheard words, brand names, or numbers. With a caption selected, open the Text style panel to change font, size, color, and outline. Whatever style you set, use Apply to all so every caption matches.

Read it out loud

Auto Captions struggles with proper nouns and acronyms. Scan the transcript once and fix anything a viewer would notice, especially the first three seconds where people decide to keep watching.

Step 5: Export

Click Export, keep the resolution at 1080p, and the captions are burned into the video. If you want a separate subtitle file instead, use the download icon on the caption track to export an SRT.

Result: a 30-second talking clip now has clean, synced subtitles that you corrected in under five minutes, ready to post to TikTok, Reels, or Shorts.