How to make word-by-word captions
By Vansh · Updated
“Word-by-word captions” can mean three different looks. All three depend on knowing exactly when each word is spoken, which is what makes them slow to do by hand in a timeline editor. With automatic word timing, they take minutes.
The three kinds
1. One word at a time
A single word fills the screen, then the next replaces it. It is the high-energy look of many YouTube Shorts and MrBeast-style videos.
- Style: One word.
- Or with any style: set Words per line to 1 in the style panel.
- Best for: fast hooks, challenges, short punchy videos. It gets tiring in long videos.
2. Words appear as they are spoken
The line builds up: each word arrives the moment you say it, by popping, bouncing, wobbling, rising letter by letter or coming into focus.
- Styles: Pop, Bounce, Jelly, Wave, Blur-in, Sticker, and Typewriter, which types each word out letter by letter.
- Best for: Reels and TikToks with a lively pace.
3. The whole line, with the spoken word highlighted
The full line is on screen and the current word changes colour, gets a box, an underline, a glow or a colour sweep, like karaoke.
- Styles: Karaoke, Box highlight, Underline sweep, Glow, Gradient sweep, and Hormozi-style with its two stacked rows.
- Best for: talking heads, tutorials and anything people should read in full.
How to make them
- Open the editor and add your video. It stays on your device.
- Transcribe it. Every word gets a start and end time; the timings are corrected so words land on the sound.
- Check the words. Tap any word to fix it. If a word appears a little early or late, nudge its start or end by 50 ms.
- Pick one of the styles above in step 4, Style.
- Set the words per line: 1 for one at a time, 2 or 3 for fast videos, up to 6 for calm styles.
- Export and post.
Getting the timing to feel right
- Lines break at natural points: at the end of a sentence, after a pause of more than 0.4 seconds, or when the line reaches its word or letter limit.
- Short gaps are bridged so captions do not flicker between lines.
- Merge words that belong together, like a name or a number with its unit, so they appear at once.
- Split long words into two if they need two beats.
- Mark key words as Emphasis. They show in the highlight colour and a little larger, and some styles shake or enlarge them.
In other editors
In most timeline editors, word-by-word captions mean splitting every caption into one-word pieces and lining each up with the sound by hand, or using a paid plugin or feature. If you edit there, you can export your video without captions and add word-by-word captions here as the last step. See the CapCut workflow for an example.



