TikTok Multilingual tool: translated ad audio and captions

The Multilingual tool in TikTok Ads Manager automatically translates an ad's audio into other languages, adds text captions for those languages, and delivers the ad to people in their device language. It is limited to Smart+ campaigns, so a campaign outside Smart+ cannot use it. Sume is a different kind of tool: its documented caption route burns captions you author or transcribe onto a finished clip, one language per job, and its docs describe no translation or dubbing route.

The TikTok facts below come from About the Multilingual tool in TikTok Ads Manager, read on 2026-10-02. The Sume facts come from the video captions, audio detach, and video inspect pages of docs.sume.com.

What does the Multilingual tool actually do?

According to TikTok's page, the tool translates the ad's audio and adds text captions for the new languages, then serves each viewer the version that matches their device language. You do not pick a language per placement; delivery follows the device.

The page does not give a language count or a list of supported languages, so check the tool inside Ads Manager before you plan a market around it.

What is outside the tool?

The page lists campaigns the tool does not serve: non-Smart+ campaigns, app retargeting, pre-registration, and Search placements. If your ad runs in any of those, the translation has to happen before upload.

The page also describes the translation as automatic. It does not describe a step where you review the translated script or pick a voice, so a brand with fixed wording needs a different route for those ads.

Where does Sume fit when the tool does not apply?

Sume's docs do not describe a translation or dubbing endpoint. What they do describe are the building blocks around it: audio detach pulls the audio track of one Sume-hosted video into a wav, video inspect can transcribe a clip with speech-to-text at $0.01 per audio minute, and video captions burns phrase-level cues onto a public video URL without running speech-to-text.

So the Sume side of a manual localization is: transcribe, translate the text yourself or with your own tool, then burn the translated cues. Each standalone caption job reserves $0.20 for a video up to 60 seconds, and one job burns one set of cues, so three languages are three jobs. For the full chain see Build an AI dubbing pipeline and Caption one video in three languages.

Which should you choose?

Use the TikTok tool when the campaign is Smart+, the objective is app promotion or web conversions, and you accept automatic output. Use a manual route when you need exact wording per market, a campaign type the tool excludes, or captions styled to your brand.

Whichever route you pick, watch the caption placement. Captions burned into the video are part of the pixels, so they travel with the file into any placement, while the tool's captions are added by TikTok. Do not stack both: a clip with burned English captions run through a tool that adds Spanish captions can show two lines of text.

Sume does not host a language picker for your ad and does not submit to TikTok Ads Manager on your behalf in anything cited here, so the finished files still go through your normal upload.

• Smart+ campaign with app or web-conversion objective: try the Multilingual tool first.
• Fixed legal or brand wording: translate the text yourself and burn cues.
• A placement the tool excludes, such as Search: localize before upload.