How to Add Captions & Subtitles to a Video (Automatically)
85% of social video is watched on mute. Captions aren’t optional anymore — they’re how most people read your video. Here’s how to add them automatically on your phone, on-device, in a couple of taps.
Quick answer
To add captions automatically in ExpoCut: select the clip with speech (a video or a voiceover on an audio row), tap Transcribe on its toolbar, choose the Standard (32 MB) or Enhanced (60 MB) on-device skill — it downloads once — and tap Transcribe Audio. In about 3 seconds the caption editor opens: the Segments tab lists every timed line (tap one to fix a word, nudge its start/end, set a font or enter/exit animation) and the Style tab previews your real words in Karaoke, Highlight, Bounce and Typewriter looks. Tap Apply and a Transcript layer lands on the timeline; captions are burned into the export.
- Transcribe sits on the clip toolbar for both video clips and audio rows — a TTS voiceover works too.
- First run downloads a speech skill (Standard 32 MB fast, Enhanced 60 MB accurate); after that it’s fully offline and nothing uploads.
- The caption editor has two tabs: Segments (timed lines, editable text, ±0.1–1 s nudges, per-segment font, RTL, enter/exit) and Style (live preview of your words).
- Style families: Karaoke, Highlight, Bounce, Typewriter — Classic Karaoke, Neon, Fire, Yellow Marker, Blue Box, Green Accent, Red Underline, Purple Gradient and more.
- Apply creates a Transcript layer (T-row) that exports burned in; Segments also has Import for an existing subtitle file.
Do it now → Auto-caption your video free — on-device
Most people watch your video without sound. Captions are how they read it — and on-screen text also boosts watch-time and accessibility. The good news: you don’t type a word. ExpoCut transcribes the speech on-device and drops in timed captions you can restyle in seconds.
Why captions matter (more than you think)
- The sound is off. Feeds autoplay muted; captions are the difference between a scroll and a watch.
- Retention. On-screen words give the eye something to track, which keeps viewers in the video longer.
- Accessibility & reach. Captions open your video to deaf and hard-of-hearing viewers and to anyone in a quiet room.
- Repurposing. A transcript is a free script you can reuse for a caption, a blog post, or a carousel.
Step 1 — Add the clip with speech
Drop your video or voiceover on the timeline
Any clip with talking works — a talking-head, a vlog, or a voiceover track. Clear audio gives the best transcription.
Select the clip
Tap it on the timeline so its toolbar appears — video clips and audio rows both have Transcribe.
Tap Transcribe and pick a skill
The Transcribe Audio sheet offers Standard (32 MB, fast) or Enhanced (60 MB, accurate). The skill downloads once; after that it works offline.
Tap Transcribe Audio
Spoken language is auto-detected. About 3 seconds later the caption editor opens with every line timed to the speech.
Your speech clip (A2)
Transcribe
Standard 32 MB · Fast
Downloads once, runs offline
It’s all on-device. Transcription happens locally — your audio never leaves the phone. That means it works offline, and private recordings stay private.
Step 2 — Fix the wording
Auto-transcription is fast but not perfect with names, slang, or noisy audio. The editor’s Segments tab lists every timed line — tap one to open it. Fix a misheard word, nudge the start or end by ±0.1, 0.5 or 1 s, split a long line with + Add Segment, or delete filler — the timing stays attached to the audio. Each segment can also carry its own font, an RTL toggle for Urdu/Arabic, and Enter/Exit animations (Fade, Slide, Slide Up…).
Timed segment — tap to edit
Nudge start / end
Font, RTL, enter/exit animation
Keep lines short. One short phrase on screen at a time reads far better than a paragraph. If a caption feels long, split it so each line lands with its beat of speech.
Step 3 — Style them for readability
Captions only work if they’re easy to read over moving footage. The editor opens on its Style tab with ready-made caption styles in four families — word-highlight Karaoke, Highlight (Marker, Box, Accent, Underline, Gradient), Bounce and Typewriter — so one tap gets you a social-ready look you can then fine-tune.
Your words, previewed live
Karaoke · Highlight · Bounce · Typewriter
Tap a style
Transcript layer on T3
The caption layer uses the same text controls as any title, so you can:
| Control | Recommendation |
|---|---|
| Font & size | A clean bold sans, ≥ 18pt. Big enough to read on a small phone. |
| Color & shade | White text on a subtle dark shade/box so it survives bright frames. |
| Position | Lower third, inside the safe area — clear of the platform’s UI buttons. Drag the Transcript layer on the canvas. |
| Highlight | Karaoke and Highlight styles punch each word as it’s spoken — the “social caption” pop. |
Step 4 — Export with the captions baked in
Captions render as a real layer, so when you export they’re burned into the video — they show on every platform, even ones that ignore separate subtitle files. Preview equals export, so what you see on the canvas is exactly what ships.
A faster workflow for social
- Cut the talking parts tight first.
- Transcribe — get the whole caption track in one tap (or narrate with Text To Voice first, then transcribe that).
- Pick a style once; it applies to every segment.
- Fix names in Segments; keep each line to one short phrase.
- Export 1080×1920 for Reels / TikTok / Shorts.
Where to go next
- Style the words like a pro: add text & titles.
- Add a narration track first: music & audio.
- Ship it right: best export settings.
Captions are the highest-leverage thing you can add to a talking video — and with on-device auto-transcription, they cost you two taps instead of ten minutes of typing.
Frequently asked questions
How do I add captions to a video automatically?
Add your clip, select it, and tap Transcribe. ExpoCut listens to the speech and generates a timed caption layer on-device — no typing. Then you can fix any words, restyle the text, and reposition it before exporting. The whole transcription runs on the phone, so your audio never leaves the device.
Are the captions burned into the video?
Yes — captions render as a real text layer, so they're baked into the exported file and show everywhere, even on platforms that don't support separate subtitle tracks. If you'd rather have selectable subtitles, you can still use the on-screen text as your script.
Do auto-captions work offline?
Yes. Transcription is on-device, so it works with no signal and keeps your audio private. Accuracy is best with clear speech and minimal background noise — if a word is wrong, just tap the caption and fix it.
Related guides
Caption your video in two taps
On-device auto-transcription turns speech into styled, timed captions — no typing, no upload. Add your clip and tap Transcribe.
Get ExpoCut Browse all guides