Most people watch video on mute, especially while commuting on the MRT or scrolling in bed. If you want your message to land, you need to add subtitles to video as standard practice, not an afterthought. This guide covers auto-caption tools, manual correction, styling, the burn-in versus sidecar decision and why captions matter for accessibility.
Why captions matter
Captions do three things at once. They keep silent viewers engaged, they help people who are Deaf or hard of hearing, and they improve comprehension for anyone watching in a second language, which matters in multilingual Singapore. Platforms also reward captioned video with longer watch times, so the effort pays back directly.
There is a small distinction worth knowing. Subtitles usually carry only spoken dialogue and assume the viewer can hear. Captions also describe relevant sounds such as “[phone rings]” or “[applause]” for accessibility. In everyday use the terms are used interchangeably, and the workflow to create them is the same.
Step one: generate captions automatically
Manually typing every word is slow, so start with an auto-caption tool and correct from there. Most editing apps now transcribe speech directly:
- CapCut generates captions in one tap and is popular for short-form video.
- YouTube Studio auto-captions any uploaded video, and you can edit the transcript in the browser.
- Premiere Pro has a built-in Speech to Text panel that creates a caption track.
- DaVinci Resolve offers auto captions in its free version.
- Descript transcribes as it imports and lets you edit video by editing the text.
Speak clearly and record clean audio to improve accuracy. The better your source sound, the less correcting you will do later.
Step two: correct the transcript
Automatic captions are a first draft, never a final one. They stumble on names, Singlish, accents, brand terms and homophones. Read every line against the audio and fix:
- Proper nouns and place names such as Bishan, Tampines or Jurong
- Numbers, prices and dates
- Punctuation, so sentences read naturally
- Awkward line breaks that split a phrase mid-thought
Aim for no more than two lines on screen at a time, and around 32 to 42 characters per line. Break lines at natural pauses so viewers can read in comfortable chunks.
Step three: style for readability
Legible captions survive small screens and busy backgrounds. Keep these principles in mind:
- Font: a clean sans-serif such as a system default. Avoid decorative fonts.
- Size: large enough to read on a phone held at arm’s length.
- Contrast: white or yellow text with a dark outline, drop shadow or semi-transparent box behind it.
- Position: keep captions clear of the bottom edge where platform buttons and progress bars sit.
Resist the urge to over-animate. Word-by-word “karaoke” highlighting suits energetic short-form content, but for interviews and explainers, steady, easy-to-read blocks work better.
Burn-in versus sidecar files
Once your captions are correct, you must decide how to deliver them. You can burn them permanently into the picture, or keep them as a separate sidecar file such as an SRT. Each has trade-offs.
| Factor | Burned-in captions | Sidecar file (SRT/VTT) |
|---|---|---|
| Editing later | Must re-export whole video | Edit text file only |
| Viewer control | Always on, cannot hide | Viewer can toggle on or off |
| Styling | Full creative control | Limited to platform defaults |
| Best for | TikTok, Reels, Stories | YouTube, streaming, LinkedIn |
| Search and SEO | Not indexed | Indexed by platforms |
| Languages | One version per language | Multiple tracks, one file each |
A common approach is to do both. Upload with a sidecar file on platforms that support it for accessibility and search, and burn in a styled version for short-form feeds where sidecar captions are not available.
Creating and exporting an SRT file
An SRT is just a plain text file listing each caption, its start and end time, and the text. Every capable editor can export one:
- Finish correcting your caption track in your editor.
- Choose Export, then select the captions or subtitles option.
- Pick SRT (or VTT for web) as the format.
- Upload that file alongside your video, or attach it in YouTube Studio.
Because it is a simple text file, you can open an SRT in any text editor to make quick fixes without touching the video itself.
Translating captions
To reach Singapore’s mixed audience, consider a second language track. Export your corrected English SRT, translate the text while keeping the timings intact, then upload it as a separate track. Always have a fluent speaker review machine translation, as tone and idiom rarely survive automatic tools cleanly.
Common mistakes to avoid
- Trusting auto-captions blindly. Always proofread against the audio.
- Tiny or low-contrast text. If it is hard to read on your own phone, it is too small.
- Captions hidden behind UI. Test where the platform’s buttons land and keep text above them.
- Cramming too many words on screen. Two lines maximum keeps reading comfortable.
- Forgetting sound cues. For true accessibility, note important non-speech sounds.
A simple workflow to adopt
Put it together into a repeatable routine. Record clean audio, auto-generate captions, correct every line, style for contrast and readability, then export both a burned-in version for feeds and a sidecar file for platforms that support it. It takes a few extra minutes per video, but it widens your audience and lifts your watch time immediately.
Start on your next clip. Turn on auto-captions, fix the errors, and add subtitles to video before you ever hit publish. Your silent scrollers, and viewers who rely on captions to follow along, will thank you.
Explore more: Basic Video Editing for Beginners · How to Make Reels and Short Videos · How to Record Good Audio for Video