For Social media teams

Turn one clip into a captioned post for Reels, TikTok and Shorts

Turn one finished clip into a captioned, vertical post for Instagram Reels, TikTok and YouTube Shorts, then into a dubbed version for a market that doesn't share your base language. Burned-in captions catch the viewers who scroll muted — most of them, on every one of these feeds — and a generated opener or a few extra seconds covers a cut that comes in short, without a reshoot. Captioning a 30-second clip costs 60 credits, $1.47 to $2.70, and that one job covers every platform you post the same file to: credits are billed per video processed, not per place it ends up. Your team still owns the review: a caption or a dub is a draft until someone checks it against what's actually playing.

The workflow

1
Decide which cuts you actually need, not just which platforms
List the platforms first — Reels, TikTok, Shorts, maybe a paid placement — then group them by the cuts that actually differ: a length under YouTube Shorts' 3-minute cap works for all three today, so one cut can often serve every platform. Credits are billed per video you process, not per platform you post it to, so the real cost driver is how many distinct cuts and languages you make, not how many feeds they go out on.
2
Fill a gap with a generated vertical clip
When the real footage is short on a hook or a cutaway, generate one from a text prompt at 4 to 15 seconds on Seedance 2.0, directly in 9:16 (or 1:1 for Shorts), in 720p or 1080p. Seedance 2.0 is the fast default; switch to Veo for a premium hero shot. At 10 credits per second, a 4-second opener is 4 × 10 = 40 credits.
Open the Video Generator →
3
Stretch a clip that comes in a few seconds short
Rather than reshoot or regenerate from scratch, continue an existing clip for 4 to 15 more seconds in the same aspect ratio and resolution. Only the Seedance 2.0 model supports extension today. At 10 credits per second, 4 more seconds is another 4 × 10 = 40 credits.
Open the Video Extend →
4
Burn in captions before anything ships
Upload the finished cut (MP4, MOV or WebM, up to 30 minutes) and leave translation off to caption it in its original language. 2 credits per second: a 30-second cut is 30 × 2 = 60 credits — one job, reused as the same file across Reels, TikTok and Shorts, since credits are charged per video processed, not per place it's posted. The Subtitler covers 20 of the Translator's 25 languages, not Dutch, Polish, Swedish, Filipino or Malay, and returns an MP4 with captions burned in, not a separate file.
Open the Subtitler →
5
Dub a version for a market that doesn't share your base language
Upload the same cut and pick one of 25 target languages, including regional variants such as Spanish for Spain or Latin America. You get back a new video with a translated, time-aligned audio track — 2 credits per second, so the same 30-second cut is another 30 × 2 = 60 credits. One job makes one language.
Open the Video Translator →
6
Caption the dub, too
Run the dubbed file back through the Subtitler with its new language set as the spoken language, so captions match what's actually playing — another 30 × 2 = 60 credits. Reused captions from the original-language version over a new dub is the most common mistake at this step.
Open the Subtitler →
7
Check it against each platform's own specs
Reels, TikTok and Shorts cap length and caption length differently, and none of these tools reframe, crop or trim for you. Work through the platform specs and the pre-publish checklist below before anything goes out.
8
Upscale only the footage that actually needs it
If the source is UGC shot on an old phone, pulled from an archive, or submitted at low resolution, upscale the finished version to 4K as the last step, after captioning and dubbing. 3 credits per second: a 30-second cut is 30 × 3 = 90 credits. Skip it for anything shot cleanly in the last couple of years.
Open the 4K Upscaler →

What one post actually costs

Live credit rates as of October 2026: captioning and translation both cost 2 credits per second, 4K upscaling 3, generated or extended video 10. A credit costs from $0.045 (Starter Pack: $9 ÷ 200) down to $0.0245 (Pro plan: $49 ÷ 2,000 a month).

Example 1: a 30-second vertical clip, captioned in English and posted to Instagram Reels, TikTok and YouTube Shorts, then dubbed into Spanish and captioned again for the same three platforms. Captioning the English cut: 30 × 2 = 60 credits. Translating to Spanish: 30 × 2 = 60 credits. Captioning the Spanish dub: 30 × 2 = 60 credits. 180 credits total, $4.41 to $8.10, for six posts — three platforms in each of two languages — because you pay once per version of the file, not once per platform it goes out on. That's 30 credits, $0.74 to $1.35, per post. 180 credits fits inside the Starter plan's 600 monthly credits: $19 for the month, or 180 × ($19 ÷ 600) = $5.70 for this batch, with 420 credits left for the next one.

Example 2: the same clip gets a 4-second AI-generated hook before the real footage starts, stretched by 4 more seconds because the first take cut off too early, and the source is a few years old so the finished cut is upscaled to 4K. Generating the hook: 4 × 10 = 40 credits. Extending it: 4 × 10 = 40 credits. Upscaling the finished 30-second cut: 30 × 3 = 90 credits. 170 credits, $4.17 to $7.65 — a one-time production cost on the source file, on top of Example 1's captioning and dubbing.

If you post in bursts, top-up packs (from $9 for 200 credits to $59 for 1,800) never expire. If you're posting every week, both monthly plans cost less per credit than any pack.

Why burned-in captions, specifically

Meta's own help center for video ads says up to 85% of Facebook video is watched without sound, and that 80% of viewers react negatively when a feed video plays sound they weren't expecting — captions exist specifically to reach that audience. TikTok's Newsroom frames its own auto-captions feature the same way: for viewers in sound-sensitive situations, an open office, public transit, who keep the sound off.

All three platforms now auto-caption by default, so the question isn't whether captions exist — it's who controls them. Burning them in yourself means one file you've reviewed and fixed before it's public, reused everywhere you post it, including outside these three apps; it also captions a dub you made yourself, which no platform's native auto-captions will ever do, since those only read the original audio.

It burns captions into the MP4 only — there's no separate SRT or VTT file. If a brief calls for an editable caption file, run the clip through OpenAI's Whisper, free and open source, on your own machine.

Instagram Reels specs to check

Meta's own help center sets Reels at 9:16, accepting anything from 1.91:1 to 9:16, at a minimum 720p and 30fps.

Length: clips can add up to 20 minutes combined, but Instagram's own discovery algorithm stops recommending Reels past 3 minutes to people who don't already follow you.
Caption: 2,200 characters on the text below the video; only the first couple of lines show before "more", so put the hook and any call to action first.
Format: MP4 or MOV, H.264 video, AAC audio, up to 4GB.

TikTok specs to check

TikTok allows far longer uploads than it's known for, but a short-form post built around sound-off scrolling is typically well under a minute regardless of the ceiling.

Length: 10 minutes recorded in-app, up to 60 minutes uploaded from a device on supported accounts.
Caption: up to 4,000 characters in the TikTok app itself; a scheduler posting through TikTok's Content Posting API is capped lower, at 2,200 — check which one your team actually uses.
Format: 9:16 at 1080×1920 recommended (540×960 minimum), MP4 or MOV.

YouTube Shorts specs to check

A video only counts as a Short if it's both vertical or square (9:16 or 1:1) and 3 minutes or under — miss either condition and YouTube treats it as a regular upload instead.

Length: 3 minutes (180 seconds) maximum, no official minimum.
Title: 100 characters, with roughly the first 50 showing in most feed previews.
Description: up to 5,000 characters, though only the first 100 or so show before a viewer taps to expand it.

Pre-publish QA checklist

Run every cut through this before it goes anywhere public, and add a native-speaker pass for any dub you're paying to promote.

Captions match the audio actually playing in that file — reused captions from a prior language version is the most common mistake.
Aspect ratio and length fit the specific platform spec above, not just "vertical" in general.
The hook and any call to action sit in the caption's first line or two, before each platform's own truncation point.
Names, numbers, prices and handles are correct, and spelled the way your brand actually spells them, in every language.
A native speaker has checked tone and meaning on every dub you're paying to promote, not just a back-translation.
The footage is actually vertical, or has been cropped to vertical in your editor — none of these tools reframe or crop for you.

Limits to know before you commit

Plan a posting schedule around these:

Generated clips run up to 15 seconds, with the exact lengths set by the model, and an extension adds 4 to 15 seconds; Video Extend supports only the Seedance 2.0 model today.
One target language per Video Translator job, and up to 30 minutes per uploaded file on the Translator, Subtitler and 4K Upscaler; Video Extend takes a 2–15 second source clip.
Neither the Translator, the Subtitler nor the 4K Upscaler reframes or crops — whatever aspect ratio you upload is what comes back, so reshape horizontal footage to vertical in your editor first.
The Subtitler covers 20 of the Translator's 25 languages: not Dutch, Polish, Swedish, Filipino or Malay.
The Subtitler burns captions into the MP4 only — no SRT or VTT file; for an editable transcript, use OpenAI's Whisper (free, open source) instead.
Neither tool translates on-screen text, stickers or graphics you've already added — redo those per language in your editor.
Generated video tops out at 1080p.

Questions

Don't TikTok, Instagram and YouTube already caption videos for free?

They all auto-caption by default now, so you're not paying for captions that otherwise wouldn't exist — you're paying for control: one file you've reviewed before it's public, reused on every platform including outside these three apps, and captions on a dub you made yourself, which no platform's native auto-captions will generate.

Does posting the same captioned clip to three platforms cost three times as much?

No. Credits are billed once per video file processed, not per platform you publish it to — captioning a 30-second cut costs 60 credits whether you post it to one feed or ten.

Can we generate clips already in the right vertical shape?

Yes. Both the Video Generator and Video Extend offer 9:16, and 1:1 for Shorts, at generation time. Pick the ratio before you generate; cropping a finished clip afterward loses quality.

What if our source footage is horizontal?

None of these tools reframe or crop it. Crop or pad it to vertical in your own editor before you caption, translate or upscale it — whatever shape you upload is the shape you get back.

Do we get an SRT or other caption file we can upload separately?

Not from the Subtitler — it delivers captions burned into the MP4, nothing separate. If a platform or a client brief needs an editable file, run the clip's audio through OpenAI's Whisper, free and open source, on your own computer.

What does a test cost?

The 20 free credits at sign-up cover a 10-second caption test (10 × 2 = 20), with no watermark. After that, a $9 Starter Pack buys 200 credits (100 seconds), and the Starter plan $19 for 600 credits a month, or 300 seconds — 5 minutes — of captions, dubbing, or any mix of the two. Failed jobs are refunded.