Editing
Video Subtitles: Why to Add Them, How to Do It, and the Styles That Hold Attention
By the ManyMotion team · · 10 min read
On social media, most videos are watched with the sound off. If your message only comes through your voice, a large part of your audience never hears it. Video subtitles fix that, but they do much more: they add rhythm, guide the eye and make your video accessible. Here's why to add them, how to do it without spending hours, and how to choose a style that holds attention instead of distracting from it.
Why caption your videos
Think about when you watch videos: on public transport, in a waiting room, at the office, at night next to someone sleeping. The sound is often off. Without subtitles, a talking-head video becomes a person moving their lips. People scroll on in a second.
Subtitles have four concrete effects:
- They make your video understandable without sound. That's the main reason, especially on TikTok, Reels, Shorts and LinkedIn.
- They create movement on screen. Words appearing in time with your voice are a constant visual change, even on a static shot.
- They improve recall. Reading and hearing the same thing at the same time helps people remember, especially a number or a name.
- They make your video accessible to deaf and hard-of-hearing viewers, and to people who aren't fully comfortable with your language or accent.
Burned-in captions or a separate file
There are two ways to display subtitles:
| Type | How it works | Pros | Cons |
|---|---|---|---|
| Burned in (open captions) | The text is part of the video image | Visible everywhere, any style, animated | Can't be turned off or edited after export |
| Separate file (closed captions) | A text file read by the video player | Can be turned off, translated, read by search engines | Style set by the platform, not supported everywhere |
On social media, burned-in captions are the norm: they show by default on every platform and can be styled and animated. On YouTube, some creators combine both: clean burned-in captions plus a separate file for search and the platform's automatic translation.
How to add subtitles to a video
By hand: precise but very slow
In a traditional editor, you have to transcribe your video, split the text into small chunks, then time each chunk on the timeline. For a two-minute video, that's easily an hour of work. And an offset of a fraction of a second is noticeable.
With automatic transcription: fast but needs checking
Most editors now offer automatic transcription. It saves a huge amount of time, but still needs proofreading: names, technical terms, mumbled words. And you still have to choose the style, the chunking and the placement.
With an AI video editor: synced word by word and placed for you
With ManyMotion, you upload your video and the AI transcribes it word by word, with the exact timestamp of each word. Captions appear right as you say each word, in the style you chose. Keywords are highlighted in color, and thanks to face detection, captions are placed away from your face. Transcription works in English, French, Spanish, Portuguese, German and many other languages.
A name transcribed wrong? Fix it in one sentence: "replace Kathryn with Catherine". All the details are on our auto captions page.
The 3 main caption styles, and when to use them
1. "Pop" style
One to three words at a time, extra bold, in the center of the screen, popping up one by one. It's the TikTok creator style. It's very readable on a small screen and creates lots of energy. The downside: on a long video it gets tiring, and it takes up a lot of space.
Best for: TikTok, Reels, Shorts, videos under a minute, high-energy content.
2. "Karaoke" style
A short line on screen, and the spoken word changes color as you say it. The eye follows the voice. It's a good balance between energy and reading comfort.
Best for: tutorials, recipes, storytelling, educational content, one- to three-minute videos.
3. "Clean" style
One or two discreet lines at the bottom of the frame, with no strong animation. It stays in the background behind the content.
Best for: LinkedIn, YouTube, training, interviews, long or corporate videos.
Readability rules to follow
- Few words at a time. Three words in pop style, a short line in karaoke, two lines max in clean style. Beyond that, nobody finishes reading before it changes.
- A thick font. Thin fonts disappear on a busy background. Bold fonts like Montserrat, Poppins, Anton, Bebas Neue or Lilita One stay readable on a small screen.
- Contrast. White text with an outline or shadow, or text on a background bar. The caption must stay readable on any image.
- The right zone. On TikTok and Reels, the bottom of the screen is covered by the caption and buttons, the top by the account name. Stay in the central area.
- Never over the face. A caption covering your eyes or mouth breaks the connection with the viewer.
- Perfect sync. Captions that run early or late are worse than no captions. The brain notices the offset instantly.
- Zero typos. A typo in your brand name or a simple word looks amateurish. Proofread before posting.
Highlighting keywords in color: getting the dose right
Coloring certain words guides the eye to what matters: a number, a benefit, a surprising word. Used well, it's one of the most powerful caption techniques. Used badly, it's noise.
- One colored word per sentence, no more. If everything is colored, nothing stands out.
- Always the same color for the same purpose. For example, yellow for numbers, green for benefits.
- A color that contrasts with your background and outfit.
- Your brand colors if you want your videos to be recognizable from the first frame.
Example request: "Add pop captions, 2 words max, Anton font, white with the numbers in yellow, centered, and cut the silences too."
Captions alongside emojis, text and zooms
In a fast-paced edit, captions share the screen with other elements: emojis popping up, hook text, section titles, numbers, zooms. The risk: an overloaded screen where the eye doesn't know where to look.
- Set a hierarchy: hook text at the top, captions in the center or bottom, emojis next to the face.
- Avoid overlap: an emoji should never cover a caption.
- Give it room: not every element needs to appear at the same time.
ManyMotion automatically places text, emojis and captions away from your face. For a complete edit (cuts, captions, zooms, music), see AI video editor, and for short videos, TikTok video editing.
The most common mistakes
- Transcribing hesitations word for word. "Ums" have no place in captions. Better yet: cut them from the video, as explained in how to remove silences from a video.
- Changing style mid-video. One font, one size, one position: keep them from start to finish.
- Skipping captions on the hook. The first seconds need to be the most readable, since that's when viewers decide to stay.
- Captions that are too small. Check on a phone at arm's length, not on your computer screen.
Captions are a key part of retention, but not the only one: read our editing techniques that keep viewers watching to the end for the hook, pacing and sound.
Frequently asked questions
How do I add subtitles to a video automatically?
+
Use an automatic transcription tool. With ManyMotion, you upload your video, the AI transcribes it word by word and places synced captions in the style you chose (pop, karaoke or clean), away from your face. You fix a misheard word by asking in the chat.
What's the best font for video subtitles?
+
A thick, simple font that's readable on a small screen: Montserrat or Poppins for a clean look, Anton or Bebas Neue for impact, Lilita One for a more playful tone. Add an outline or shadow for contrast.
How many words should I show at once?
+
One to three words in pop style for short videos, a short line in karaoke, two lines max in clean style. Beyond that, viewers don't have time to read.
Can the subtitles be translated?
+
With ManyMotion, captions follow the language spoken in the video: transcription works in many languages, but the tool doesn't translate or dub your voice.