How YouTube auto dubbing works, and its limits
What YouTube auto dubbing is, how it works in 2026, which languages it covers, and the lip sync gap that still makes auto-dubbed videos look dubbed.
More than 6 million people a day watched at least 10 minutes of auto-dubbed YouTube content as of December 2025, per YouTube’s February 2026 announcement. YouTube auto dubbing is a free feature that automatically generates translated audio tracks for your videos, so a viewer in Brazil hears your upload in Portuguese without you doing anything. What it does not do is touch the picture: the dub is audio-only, so the speaker’s lips keep moving in the original language. That last step is what sync. labs adds. It generates a dub in the creator’s own cloned voice and re-syncs the lips to match, so the track you publish looks shot in that language instead of spoken over it.
This page covers what auto dubbing is, how it actually works in YouTube Studio, which languages it supports, and how to decide when the automatic version is enough.
YouTube auto dubbing creates translated audio tracks automatically
Auto dubbing detects your video’s language, generates dubbed audio in other languages, and attaches each dub to the same video as a separate audio track. Viewers get the track matching their language settings, and the video is marked “auto-dubbed” in the description so nobody is misled about what they are hearing.
The feature rolled out in stages. YouTube launched it in December 2024 for hundreds of thousands of knowledge and information channels in the Partner Program, dubbing English into French, German, Hindi, Indonesian, Italian, Japanese, Portuguese, and Spanish. In February 2026 it opened to everyone, alongside an expressive speech upgrade, built with Google DeepMind and Google Translate, that carries the original delivery’s pitch and intonation into the dub. It costs nothing and is on by default for eligible channels.
Auto dubs appear in YouTube Studio before or after they publish
When you upload a video, YouTube generates the dubs on its own and lists them in the Languages section of YouTube Studio, desktop only. From there the mechanics matter, because they shape what you can and cannot fix. Per YouTube’s help center as of July 2026:
- You cannot edit an automatic dub. If a line is mistranslated or a name is mispronounced, your options are to unpublish the dub or replace it with your own track. There is no line-level correction.
- You can review before anything goes live. Turning on “Publish manually” in channel settings holds every dub for review instead of publishing it automatically.
- Videos over 120 minutes are ineligible, as are videos with no speech, music-only audio, speech paced too fast, an undetectable source language, or an active Content ID claim.
- Quality varies with the source audio. YouTube notes that accents, dialects, background noise, proper nouns, idioms, and jargon all produce errors.
- You can turn it off at any time under Settings, then Content, then Automatic dubbing in YouTube Studio, and YouTube says leaving it on has no negative effect on how your original video is recommended.
Auto dubbing covers about 30 languages in 2026
Language support depends on direction. Dubbing into English works from 29 source languages, while dubbing from English works into 20 languages, per YouTube’s help center as of July 2026. Expressive speech, the more natural-sounding voice mode, covers 8 languages so far: English, French, German, Hindi, Indonesian, Italian, Portuguese, and Spanish, per YouTube in February 2026.
| Direction | Coverage as of July 2026 |
|---|---|
| Any supported language into English | 29 source languages |
| English into other languages | 20 languages |
| Expressive speech (natural delivery) | 8 languages |
For comparison, a dedicated dubbing tool is not bound to YouTube’s language pairs: sync. labs dubs across 95+ languages in any direction, and the output is a normal video file you can publish anywhere, not only as a YouTube audio track.
Auto dubbing does not move the speaker’s lips
An auto dub changes the sound and leaves the mouth speaking the original language, which is why an auto-dubbed video still reads as dubbed the moment a face is on screen. Viewers catch mismatched lips faster than any translation error. YouTube knows this is the missing half: the same February 2026 announcement says a lip sync feature is being piloted to match mouth movements to the dubbed audio, though as of July 2026 it has not shipped to creators.
The voice is the other gap. Expressive speech carries your tone and energy, but the dub is still a generated voice reading a translation, not your voice. For a channel built on a person, those two things, the mouth and the voice, are most of what makes a translated video feel native.
This is the specific problem sync. labs solves today. sync-3 reads the whole scene and re-times the speaker’s mouth to the new audio, up to 4K at 60fps, holding on side profiles, low light, and multi-speaker shots. Voice cloning generates the translated speech in the creator’s own pitch, tone, and cadence across 95+ languages, and the whole thing runs as one pass: translation, voice, and lip sync together, usually in under three minutes.
When auto dubbing is enough, and when to bring your own dub
Auto dubbing is a genuinely good default for a free, zero-effort feature, and for some formats it is all you need. The decision comes down to how much your video depends on a face and a voice.
| Your format | Best fit | Why |
|---|---|---|
| Screencasts, tutorials, voiceover-only videos | YouTube auto dubbing | No face on screen, so the lip sync gap never shows |
| Testing whether a language has an audience | YouTube auto dubbing | Free signal before you invest in real localization |
| Face-to-camera videos | Your own dub with lip sync | Mismatched lips are the first thing viewers notice |
| A channel built on your voice and delivery | Your own dub with a cloned voice | A generated narrator breaks the thing people subscribed to |
| Publishing beyond YouTube (TikTok, Reels, courses) | Your own dub | Auto dubs exist only as YouTube audio tracks |
YouTube supports both paths on the same video: creators keep full control and can replace automatic dubs with their own tracks. The workflow is to translate the video with sync. labs, then upload the result as a multi-language audio track in YouTube Studio. For a wider look at the tradeoffs between dubbing approaches, see AI dubbing vs subtitles vs voiceover.
Frequently asked questions
What is YouTube auto dubbing?
YouTube auto dubbing is a free feature that automatically generates translated audio tracks for eligible videos and attaches them to the same upload. Viewers hear the track matching their language settings, and the video is labeled auto-dubbed in the description.
How do I turn off auto dubbing on YouTube?
In YouTube Studio, go to Settings, then Content, then Automatic dubbing, and switch the feature off. You can also keep it on but select Publish manually, which holds every dub for your review before it goes live.
Does YouTube auto dubbing include lip sync?
No. Auto dubbing changes only the audio, so the speaker's lips keep moving in the original language. YouTube said in February 2026 that it is piloting a lip sync feature, but as of July 2026 it has not shipped. Tools like sync-3 re-sync the lips to the dubbed audio today.
Which languages does YouTube auto dubbing support?
As of July 2026, YouTube dubs into English from 29 source languages and from English into 20 languages. Expressive speech, the more natural voice mode, covers 8 of those languages.
Can I edit a YouTube auto dub?
No. Automatic dubs cannot be edited, only unpublished or deleted. If a dub has errors, the fix is to generate your own dubbed track with a tool you control and upload it as a multi-language audio track on the same video.
If your videos have a face on camera, make the dub match it: run one through sync. labs, pick a language, and compare the result against the auto dub on your own footage. The first three videos each month are free.