Feature
Word-timed captions, rendered into the file itself
Most captioning tools sit downstream of the edit. Here the captions come out of the same transcription that drives the cut. They land on the beat of your speech rather than a syllable behind it. Here is where the timing comes from, which of the six packs to pick, and what emphasis is for. By the end you will know why captioning adds nothing to the bill.
- Caption and style packs
- 6
- Extra credits to caption
- 0
- Words worth emphasising
- 1-2
Timed from your audio, not estimated from your script
The captions come from a transcription of the take you uploaded, with word-level timing. Your words, in your order, at the moment you said them.
Nothing is paraphrased. Nothing is summarised into a tidier sentence than the one you spoke.
The timings come from the audio rather than from an estimate, so words land on the beat of your speech. Trailing a syllable behind is the tell of cheap captioning.
They render into the picture on export, so they are part of the frame that goes out. On a feed, a separate subtitle track is often ignored and often never loaded.
All of that sits inside the 100 credit batch, alongside the directing pass and the footage search. Only the export is charged again, at 20 credits per finished minute.
From your audio to a burnt-in caption
Every step here is inside the 100 credit batch. Restyling afterwards costs nothing.
Transcribe
Word-level timing
Pick a pack
One of six
Mark emphasis
One or two words
Render in
Into the 9:16 MP4
Six packs, and the wrong one reads as hesitant or as shouting
You pick from six caption and style packs rather than assembling a look from scratch. Each is a considered combination of typeface, weight, size, colour and rhythm.
Within a pack you can change colour, weight, size and position. What you cannot do is build a look word by word, and that limit is deliberate.
The fastest way to make an ad look amateur is a font decision made at eleven at night by somebody who wanted to be finished.
Pick the pack against the ad rather than against your brand guidelines. A calm editorial look on a hard direct-response hook reads as hesitant. Bold-impact under a considered founder read reads as shouting at a friend.
Switching packs is one action and costs nothing, so test two. Export both, watch them on a phone, keep the one you read without effort.
- clean-ugc, for a real person talking to camera
- bold-impact, for a hard hook that has to stop a thumb
- karaoke-flow, for reads with rhythm and momentum
- editorial, for a calmer, more considered tone
- fast-native, for high-pace cuts that match feed behaviour
- minimal-lux, for premium products where restraint is the signal
This is the whole editor
Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.
One emphasised word is a headline. Five is somebody shouting
You can emphasise a word so it pops in the captions. Used properly this is the highest-return action in the editor.
In a muted feed the emphasised word is your headline. It does the work a voice would have done with stress.
Put it on the word carrying the claim. The number. The timeframe. The thing that is free. The thing that differs from what they use now.
One or two per ad, no more. Emphasise five words and you have emphasised none of them. The ad then reads as somebody shouting at a stranger.
Above all of it sits the opening headline, which you write or pick from drafts. That line has to survive the first half second, before anybody has heard a syllable.
There is one more reason to spend the emphasis carefully. On a phone at arm's length a viewer reads about three words at a glance. The emphasised one is what they take away when they scroll on mid-sentence.
Two emphasised words in a sixty word read
Emphasis works by contrast. The fifth emphasised word cancels the first one out.
What a dedicated captioning tool wins, and the five rows it cannot reach
A dedicated captioning tool wins two rows and both are real. It captions an edit you have already finished, and it translates into another language.
Neither is what an ad needs. An ad needs captions built from the same transcript as the cut, and a tool sitting outside the edit cannot do that.
Transcription accuracy tracks recording quality, so a windy street produces more mistakes than a quiet carpeted room. Fixing a mis-heard word fixes the caption and the footage search in the same action.
Captions come out in the language you spoke, because this is a transcription workflow. One take gives you one language.
The typeface comes from one of six packs. Colour, weight, size and position are yours to set inside whichever pack you pick.
The scope is one take of up to three minutes exporting 9:16. That is the shape the captions are positioned for, clear of the username and the button column.
Captioning inside an ad maker against a captioning tool
Two rows go to the dedicated tool. The five under them are what a tool outside the edit cannot reach.
| A dedicated captioning tool | Cutroom | |
|---|---|---|
| Captioning an edit you already have | Point it at any finished file | It builds the ad from your take |
| Translating into another language | Usually a paid add-on | One take gives you one language |
| Word-level timing with no pass by hand | Depends on the tool and the audio | Inside the 100 credit batch |
| Footage placed from the same transcript | Captions only | Highlight a phrase, a clip covers it |
| Fixing a word improves the edit, not the caption alone | The caption changes, nothing else does | The footage search reads the correction |
| Positioned clear of the platform chrome | Wherever there was room in the frame | Set for one shape, clear of the buttons |
| Included in the price of the edit | A second tool with its own bill | No credits beyond the 100 credit batch |
Restyling six times costs nothing, because only the download is charged
Switching packs, changing colour and position, and emphasising a word are all free and repeatable. Only the export is charged, at 20 credits per minute of finished video.
So a thirty second ad, captioned and downloaded, is 110 credits all in. Six restyles of that same ad before you export cost nothing at all.
The 7 day trial gives you 300 credits with no card. Basic is $39.99 a month for 2,500 credits, Premium is $79.99 for 5,000, and a $15 top-up adds 1,000.
One consequence is worth planning for. The captions come from what you said, so the read has to be written for reading. Long clauses look fine spoken and look like a wall of text on a phone.
Write in short sentences with the claim near the front. The captions then do half the selling before anybody turns the sound on.
Two exports of the same ad in two packs is 20 credits. That is the cheapest opinion you can buy about your own captions.
Questions people ask
- Are the captions burned into the video?
- Yes. They render into the exported 9:16 MP4, which is what you want on feeds where a separate subtitle file is often ignored or never loaded.
- How do I fix a word it heard wrong?
- Correct it in the transcript. That updates the caption and also updates what the machine understands the line to be about, which feeds the footage search.
- Can I change the caption style after the edit is done?
- Yes. Switching packs is a single action and the whole ad is rebuilt with the new look. Colour, weight, size and position are adjustable within a pack.
- Do captions cost extra credits?
- No. They are included in the 100 credit batch along with the directing pass and the footage search. Only the export is charged, at 20 credits per finished minute.
- Is it right for every job?
- It is built for captions that come out of the same transcript as the cut. If you need to caption a webinar you already edited, or a second language, use a dedicated captioning tool for that file and bring the ads here.
Assume the sound is off. The words land on the beat you said them, burnt into the file, for no credits beyond the batch.