Dubbing is quoted per finished minute, per language, and the range is wide enough to be useless as a number on its own. What decides where you land in that range — and how to tell whether a language pays for itself — is straightforward once you know what is actually being bought.
The floor is free, and that matters
YouTube’s own automatic dubbing costs nothing and is switched on by default for eligible channels. Any paid dub is therefore not competing with silence; it is competing with a free synthetic track that already exists. That is the right frame for the money question: you are paying for the difference between a machine reading your video and a person performing it. On channels where the delivery is the product, that difference is the product. On videos that are mostly information, it often is not. We set out where the line falls in auto-dubbing versus a recorded voice.
What you are paying for, line by line
- Translation and adaptation. Not a subtitle translation — a script that carries your meaning and fits the time the original line occupies, which usually means rewriting it. Languages expand: the same sentence in Spanish or German is routinely longer than in English, and someone has to make it fit without sounding rushed.
- The voice. Casting a performer who can carry your register, then the recording session itself. This is the largest line and the one that scales with quality.
- Direction and retakes. Reviewing takes before anything is mixed, so a flat read does not reach your channel.
- The mix. Sitting the new voice into your original music and effects, at your video’s exact length, matched in loudness to the original.
- Verification. Checking the delivered file against the published video, so the track is not removed later — the failure that costs more than the dub.
- Upload and packaging. Putting tracks on the videos and translating titles and descriptions, which is what makes a dub findable rather than merely present.
What moves the price
| Factor | Effect |
|---|---|
| Speech density | A talking-head explainer is more words per minute than a build video. Words are the unit of work, not minutes. |
| Number of speakers | Every distinct voice in the video is another performer to cast and record. |
| Language | Voice rates differ by market. Japanese and German usually cost more than Spanish or Portuguese. |
| Source materials | A script or subtitle file, and separate music and effects stems, cut the work. Nothing but a finished video adds to it. |
| Volume | A steady weekly pipeline is cheaper per minute than one-off videos, because casting and setup amortise. |
| Turnaround | Rush costs more everywhere. Our normal pace is 3–4 days per video, 1–2 on rush. |
How to tell whether it paid
One calculation, from your own exports, once a month:
- From the Geography report, take estimated revenue and engaged views for the countries that speak the language, and work out revenue per 1,000 engaged views there.
- From the Audio track report, take the engaged views on that dubbed track.
- Multiply. That is what the dub earned that month.
- Compare against what that language cost. Dubs keep earning on old videos, so the comparison is cumulative, not monthly.
Most channels reach payback on a language over months, not weeks, and evergreen videos get there first — which is why testing on your best-performing back catalogue beats testing on this week’s upload. Choosing the language is the same arithmetic run forwards.
Getting a quote
Every video is different: its length, how much of it is speech, how many speakers, and which languages you want. Tell us about your channel and we will talk it through with you.
The first minute is free for serious inquiries — one of your videos, in a language you choose, so a quote is attached to something you have heard. Request a sample.
Related: how to add the track · when the option is missing.