The short answer: for most paid-social video ads, you don't need a separate closed caption file at all — but if a platform or workflow does call for one, SRT (SubRip) is the safest closed caption file format for video ads, with WebVTT as the second most accepted option. Both are plain-text, widely supported, and easy to edit. Formats like SCC or TTML belong to broadcast and streaming workflows, not Reels, TikTok or Meta ad campaigns.
This guide breaks down the actual caption formats, when you need a file versus burned-in text, and how each major ad platform treats captions — so you stop guessing and start shipping ads that are watchable on mute.
What is a closed caption file, exactly?
A closed caption file is a separate text file, timed to the video, that a video player displays as an overlay — and that the viewer can usually turn on or off. This is different from open captions (also called burned-in or hard-coded captions), which are rendered permanently into the video frames and can't be switched off.
For most ad formats, a closed caption file is optional metadata that sits alongside the video upload. For a native vertical ad on TikTok or Instagram, the platform's own auto-captioning tool — or your own burned-in text — usually does the job without a file upload at all.
The main closed caption file formats for video ads
If you do need to produce or upload a caption file, you'll run into one of a handful of formats. Here's what each one actually is and where it's used.
| Format | Extension | Where it's used | Notes for ad creators |
|---|---|---|---|
| SRT (SubRip) | .srt | Social platforms, editing software, most ad managers that accept captions | Simplest format: numbered blocks of timecode + text. Easiest to edit by hand. |
| WebVTT | .vtt | Web video players, some social upload tools | Similar to SRT but web-native; supports basic styling and positioning. |
| SCC (Scenarist Closed Caption) | .scc | Broadcast TV, some streaming platforms | Rare in paid social; built for broadcast caption standards, not needed for feed or Reels ads. |
| TTML / DFXP | .ttml / .dfxp | Streaming platforms, some CMS-driven video players | XML-based, more complex; overkill for a 15–30 second ad. |
| Burned-in / open captions | N/A (part of the video file) | TikTok, Reels, Shorts, Stories, in-feed ads | Not a separate file — text is rendered into the MP4 itself. Works regardless of platform settings. |
For nearly every paid-social use case, you'll either work with an SRT file or skip files entirely and burn captions in. VTT shows up occasionally when a platform's upload tool specifically asks for it. SCC and TTML are worth knowing exist, but you're unlikely to ever need them for a vertical ad.
Closed caption file vs burned-in captions: which should your ad use?
This is the real decision most advertisers face, and it matters more than the file format itself.
- Burned-in captions win on reach. They show up the same way for every viewer, on every device, whether or not the viewer has captions enabled in their settings. For an ad designed to work on mute, this matters a lot.
- Caption files win on accessibility and editability. A separate file can be translated, resized, or turned off by the viewer, and it's the standard expected for long-form or on-demand video content.
- Ad length changes the calculus. A 15–30 second ad is short enough that burned-in, word-by-word captions rarely feel intrusive — and they remove any dependency on the viewer's caption settings being on in the first place.
- Platform auto-captions are a third option. Several platforms will auto-generate captions from your audio track once uploaded. These are convenient but inconsistent in accuracy, so many advertisers still prefer to bake in reviewed, on-brand captions instead.
For short-form ad creative specifically, the practical move for most teams is to burn captions directly into the video and skip the caption file altogether. Treat an SRT or VTT file as something you'd produce only if a specific platform, client, or accessibility requirement asks for it — test whether your target placement even supports caption uploads before building that step into your workflow.
How different ad platforms handle captions
Caption support varies by platform and by placement, so check the current upload options before assuming a caption file will be used. As a general pattern:
| Platform | Caption file upload? | Auto-captions available? | Common approach for ads |
|---|---|---|---|
| TikTok | Limited, placement-dependent | Yes, auto-generated on upload | Burn captions into the creative for consistent styling |
| Meta (Facebook/Instagram) feed & Reels ads | Supported for some video ad formats | Yes, for several placements | Mix of auto-captions and burned-in text, depending on brand style |
| YouTube Shorts / in-stream ads | Supported via creator/ad tools for some formats | Yes | Burned-in captions preferred for short vertical ads |
| LinkedIn video ads | Supported | Limited | Caption file useful for longer, information-dense B2B videos |
Because support changes by placement and account type, treat any specific percentage or exact spec you read elsewhere as something to verify in your ad account at the time of upload, not a fixed rule.
How to create a closed caption file for a video ad
If your workflow genuinely calls for an SRT or VTT file — for example, a client wants captions that can be toggled off, or you're repurposing the ad for a platform that requires uploaded captions — here's the simple path:
- Transcribe the audio. Use your editing software's auto-transcription feature, a dedicated captioning tool, or do it manually for a short ad script.
- Time each caption line to the audio. Most tools will auto-sync; double-check timing on key words like the offer or call to action, since misaligned captions on a short ad are easy to notice.
- Export as SRT (or VTT if the platform requests it). Keep line length short — a handful of words per line reads faster than a full sentence on a small screen.
- Review for accuracy on mobile. Watch the ad at actual vertical ad size, not full-screen on a desktop monitor, since caption size and line breaks look different at 1080×1920.
- Keep a master copy. Store the caption file alongside your video export so you can reuse or translate it later without re-transcribing.
If instead you're burning captions directly into the video — the more common path for short vertical ads — you still want the same discipline: short lines, synced tightly to speech, and legible against your background at mobile size. This matters just as much as getting the underlying ideal file format for vertical video ads right, since a caption that clips off the frame edge undoes the benefit of captioning at all.
Common mistakes with caption files and captioned ads
- Assuming every placement accepts an uploaded caption file. Many short-form placements don't; check before building your workflow around a file that never gets used.
- Relying only on platform auto-captions without reviewing them. Auto-generated captions can misread product names, numbers, or offer details — always proofread before launch.
- Making captions too small or too low-contrast. A caption that's hard to read in the first second of a scroll-stopping ad defeats the point of captioning at all.
- Forgetting captions when localizing an ad. If you're running the same ad across languages, the caption file (or burned-in text) needs to be translated too, not just the voiceover.
- Ignoring audio sync when re-editing. If you trim or re-order a video after exporting a caption file, the timecodes break — always re-sync captions after any edit to length or pacing, which also matters if you're adjusting against video length guidelines for feed ads.
How FrameNotion handles captions
FrameNotion skips the caption-file question entirely by building word-by-word captions directly into every ad it generates. You paste a product link, and FrameNotion AI writes and renders a 30-second vertical ad with an AI voiceover, music cut to the beat, and burned-in captions synced to the speech — so the ad works whether the viewer has sound on or off, with no separate SRT file to manage. You can also choose music-only or silent versions if you don't want voiceover and captions at all.
Because the captions are generated as part of the ad rather than uploaded separately, there's nothing to re-sync if you request a change: FrameNotion takes text and color edits after the ad is made, without starting from scratch. Every ad is also delivered in 4:5, 1:1 and 16:9 alongside the original 9:16, so the same captioned creative is ready for feed, Stories and widescreen placements without re-exporting captions for each ratio. You can see finished examples, captions and all, on the examples page, or read more about the process on the features page.
Frequently asked questions
Do I need a closed caption file for a TikTok or Reels ad?+
Usually not. Most short-form placements either auto-generate captions from your audio or let you burn captions directly into the video. A separate SRT or VTT file is more common for longer-form or web-hosted video than for short vertical ad placements.
What's the difference between SRT and VTT?+
Both are plain-text, time-coded caption formats. SRT is the older, more universally supported format and the safer default. VTT is web-native and supports some basic styling, and shows up when a specific upload tool requests it.
Can I convert an SRT file to burned-in captions?+
Yes. Most video editing tools can import an SRT file and render the text directly into the video frames, which turns a file-based caption into an always-visible open caption — useful if you've already transcribed the ad but want captions that work regardless of viewer settings.
Will auto-generated captions work well enough for an ad?+
They can work, but always review them before launch. Auto-captions sometimes misread product names, prices or offer details, which is risky on a short ad where there's no time to recover from a confusing caption.
Does caption file format affect accessibility compliance?+
It can. Formats like SRT and VTT are widely recognized by accessibility standards because viewers can toggle them on or off, which burned-in captions cannot. If accessibility compliance is a requirement for your campaign, confirm the specific standard you need to meet rather than assuming burned-in text satisfies it.