Skip to content

Guide

TikTok will caption the video itself. Here is the route

TikTok has generated captions built in, they cost nothing, and for most people they are genuinely good enough. This walks the in-app route step by step, then makes the case for burning captions in from an editor instead, which is a real decision with a real cost on both sides.

What the built-in caption tool costs you
$0
Steps from an uploaded take to captions on screen
6
Words a caption card should hold at a time
2-4

The in-app route, in six steps

The built-in tool is the right answer for most organic posting and it is worth saying that before anything else on this page. It is fast, it produces a caption track a viewer can switch off, and it does not add a processing step to your file.

The exact wording of the buttons moves between app versions, so treat the labels below as landmarks rather than a script. The shape of the flow has been stable for a long time.

Give the transcript one proper read before you post. That single pass is where almost all the value of this route sits, and it is the step people skip.

  • Record or upload the video, then confirm to move on to the editing screen.
  • Open the text tools down the right hand side and choose Captions. On some versions it sits behind a menu marked with three dots.
  • Wait a few seconds for the transcript. It runs on speech recognition, so ordinary sentences come back well and names come back badly.
  • Tap to edit and fix every wrong word now. A mis-heard brand name is the one error viewers always notice.
  • Set the style and drag the block clear of the bottom band, where the caption, the handle and the buttons sit over the picture.
  • Post. Viewers can turn the captions off from the share sheet, which is a feature rather than a flaw.

Fix the words. This is the step people skip

Automatic transcription is very good at ordinary speech and reliably wrong on the words that matter most to you: your brand, your product names, and the technical terms your customers actually search for.

Read the whole transcript rather than the first two lines. A caption that says the wrong ingredient where you said the right one is worse than having no captions at all, because it is confidently wrong on screen for two seconds and somebody will screenshot it.

Watch numbers and units as closely as names. Prices, dosages, measurements and dates are the second most common failure, and they are exactly the parts a viewer pauses on.

One reading pass costs about the length of the video. It is the highest return minute in the entire process, and it is free in the only sense that matters, which is that it needs no tool.

This is the whole editor

Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.

CLIPS · 5I have thisexact conversationeverysingle week. Somebody sits down and says,oh yeah, I takecinnamonevery day.And honestly, doc, I have no idea if it works.So let me tell you what isin that capsule.a clip lands on these wordscut from the editTAKING CINNAMONEVERY DAY?is it doing anythingHeadlineMusicCaptionsTHIS VIDEOLength25.0sClips5Words removed18Export video

When burned-in captions from an editor are worth the extra step

Burned in means the words are part of the picture, drawn into the frame before the file leaves your editor. The viewer cannot switch them off and the platform cannot restyle them, which is both the benefit and the cost.

There are five situations where that trade lands in your favour, and outside them the built-in tool wins on effort alone.

The cost, stated plainly: burned-in words cannot be turned off, cannot be auto-translated by the platform, and do not feed the caption layer that some accessibility settings read. You are buying control and paying in flexibility, and it is worth knowing which one you actually need.

If you want to see what different caption looks do to the same sentence before committing to one, /tools/caption-style-preview plays a line you type in five styles in the browser, with no account needed.

  • You post the same file to more than one platform and want one consistent look across all of them.
  • You run the video as an ad, where the in-app caption tools are not part of the upload flow.
  • You want specific words emphasised, sized or coloured, which the built-in tool will not do.
  • You are handing the file to somebody else, a client or a media buyer, and it has to arrive finished.
  • The caption style is part of your brand, and letting each platform restyle it undoes the point.

Built-in captions against burned-in captions

Neither column is the right answer. The question is which of these six rows you actually care about.

TikTok's built-in captionsBurned in from an editor
What it costsNothingWhatever the editor costs
Viewer can hide themYes, from the share sheetNo, they are the picture
Platform auto-translationPossibleNot available
Word-level emphasisLimitedFull control
Same look everywhereNoYes
Best forMost organic postingAds, multi-platform, client work

The accessibility argument, which applies to both routes

Captions began as an accessibility feature and that is still the most important reason to use them. Viewers who are deaf or hard of hearing need them to receive the video at all, and treating that as an optional extra is a choice worth naming out loud.

The overlap with everybody else is enormous. A big share of feed viewing happens with the sound off, in offices, on public transport, and next to sleeping people, so captions are the default reading mode for a great many viewers who can hear perfectly well.

Some habits help both groups at once. Keep two to four words on screen at a time so the eye takes a card in one glance. Keep contrast high, with solid or outlined type rather than thin text over busy footage. Keep the block out of the bands where the interface sits.

Do not caption only the dialogue. If something important happens in the audio and not in the picture, say it in the video, because a caption track made purely of speech leaves out whatever the sound was carrying.

And correct the transcript, always. An automatic caption with errors in it is an accessibility feature that has stopped working for exactly the people who cannot check it against the audio.

Use the built-in tool, unless you are doing this every week

If most of your posting is organic and lives on one platform, use the built-in captions, fix the words, and spend nothing. That is the honest recommendation on this page and it is not a set-up for the paragraph after it.

Where an editor earns its place is volume and consistency. Cutroom transcribes your take, trims the silence, drops b-roll on exactly the words you point at and burns word-timed captions into a finished 9:16 MP4 in one of six styles, so every post looks the same without you rebuilding it each time. The batch is 100 credits and the export is 20 credits an output minute. The first three are free in your first seven days, with 600 credits and no card, and carry a Cutroom mark; Lite at $19.99 a month for 1,250 credits is about ten finished one-minute videos, cancelled in one click.

The limits worth knowing before you decide: the captions come out burned into the picture, there is no subtitle file to download, the output is vertical MP4 only, and the source take cannot run past three minutes. If a downloadable caption file is what you actually need, the in-app route is closer to it than we are.

Questions people ask

Are TikTok's automatic captions accurate?
Good on ordinary speech and unreliable on brand names, jargon, numbers and anything said over music. Read the transcript before you post. That pass is where the quality comes from, not the tool.
Can viewers turn the captions off?
Yes, from the share sheet, which is one of the strongest arguments for the built-in route on organic posts. Words drawn into the picture by an editor cannot be turned off by anybody.
Do captions get a video more views?
They help retention, which is the thing actually being measured, mostly because a muted viewer can follow the video at all. Nobody can honestly promise a view count from adding captions.
Should I use built-in captions and burned-in ones together?
No. Two sets of words on screen at once is unreadable, and the platform will happily draw its own over yours. Pick one route per video and stay with it.
Can I get a subtitle file instead of captions in the picture?
Not from us. Our captions are drawn into the frame and there is no separate file to download. The in-app tool is closer to what you want, and dedicated captioning apps will export a subtitle file directly.

If you post to one platform and you are not in a hurry, the built-in captions are the right answer and they cost you nothing. For the same look on every post without redoing it by hand, the first three finished verticals are free in your first seven days and $19.99 a month covers around ten after that.

3 videos free, no card3 finished videos free in your first 7 days, no card. They carry a Cutroom mark; Lite at $19.99/month removes it