How to make TikTok-style word-by-word captions
Last updated 13 August 2026
The caption style that runs through TikTok, Reels and Shorts is not a subtitle track — it is a few words on screen at a time with the current word lit up as it is spoken. It needs word-level timings, a font heavy enough to survive compression, and placement that dodges the platform's own buttons. YouClip generates the timings from the audio and gives you the treatment, font, colour and position as controls you set once and reuse.
Step by step
Add the clip and let it transcribe
Import the video. YouClip transcribes it with a timing per word, which is what a karaoke caption needs — a line-level subtitle cannot highlight one word at a time.
Pick a highlight treatment
Choose how the active word stands out: recolour it, wipe a highlighter bar behind it, underline it, outline it, or scale it up as it is spoken. Each reads differently on a busy feed, so try two on the same clip before committing.
Set the font
Use any Google font or install your own. A heavy, wide sans with a strong outline survives compression and small screens; a thin display font disappears the moment the platform re-encodes your upload.
Choose colours that survive the background
White text with a dark outline works on almost any footage. Save the accent colour for the highlight itself, so the eye tracks the word being spoken rather than the whole line.
Place the block on the preview
Drag the caption where you want it on the preview, which shows exactly what will be exported. Keep it clear of the buttons down the right edge and the platform's own caption bar at the bottom.
Add emphasis and emoji if they fit
Emphasise individual words in a line so the key phrase hits harder, and turn on auto-emoji to append a keyword-matched emoji per line. Both are easy to overdo — one strong emphasis per line beats five.
Save it as a preset, then render
Store the font, colours, treatment and position as a preset so every future clip matches without redoing the setup. Export at 1080x1920 with the captions burned in.
The five highlight treatments
- Recolour — the spoken word changes colour. The cleanest option and the safest on busy footage.
- Highlighter wipe — a coloured bar sweeps behind the word. Loud, and very legible over messy backgrounds.
- Underline — a rule tracks along under the line. Subtle; good for talking-head clips with a calm background.
- Outline — the active word takes a contrasting border. Works when the caption colour already matches your brand.
- Scale — the word pops larger as it lands. The most energetic; best on fast delivery.
Readability beats decoration
Captions are read in motion, on a small screen, often at a glance. That means fewer words per line, high contrast against whatever is behind them, and a size big enough to read without leaning in — usually larger than feels right on a desktop preview. Check the clip at phone size before you render; a caption that looks bold on a laptop can be marginal on a handset.
Emphasis and emoji, used sparingly
Emphasising a word inside a line pulls the eye straight to the point of the sentence, which is useful once per line and noise three times per line. Auto-emoji appends a keyword-matched emoji to a caption line — great on casual, high-energy content, wrong on anything that needs to read as authoritative. Both are switches, so you can set them per clip or bake a decision into a preset.
Take it further
- How automatic captioning works from import to export.
- Crop a landscape clip to 9:16 before you place the captions.
- Apply one caption style to a whole folder in a single render queue.
- Caption in another language to reach a second audience.
Frequently asked questions
What are karaoke captions?
Captions where each word highlights at the moment it is spoken, instead of a static line of subtitles sitting on screen. They need word-level timings from the transcript, which is why they cannot be built from a normal SRT subtitle file.
Why do TikTok captions work so well?
Most feeds are watched on mute or half-attention, and a word lighting up in time with the audio gives the eye something to follow. It keeps viewers on the clip through the middle, where short-form videos usually lose them.
Should I use the platform's own auto-captions instead?
Platform captions are fine for accessibility but they all look the same, they sit where the platform decides, and they do not carry your font or colour. Burned-in captions are part of your channel's look and render identically wherever the file is uploaded.
What font should I use for captions?
A bold, wide sans-serif with a heavy outline or shadow. It has to stay readable at phone size on top of moving footage and after the platform re-compresses your upload. YouClip takes any Google font or a font file of your own.
Where should captions sit in the frame?
Centred horizontally, roughly in the lower-middle third — above the platform's own caption and handle overlay, and clear of the button column on the right. The preview shows the exported frame, so you can position it exactly rather than guessing.
Can I reuse one caption style across every clip?
Yes. Save the whole look as a preset and apply it to any clip or to a whole bulk render. Presets can also be exported as a share code, so a team posts in one consistent style.