ElevenLabs Video

Positioning

ElevenLabs built its reputation on text-to-speech, and its dubbing product — Dubbing Studio and the newer Dubbing v2 automatic pipeline — is the clearest attempt by anyone to make translated video sound like the same person speaking another language. Our catalog names this entry "ElevenLabs Video", but what actually ships is the dubbing line: an end-to-end pipeline that takes audio or video, transcribes it, translates it, synthesizes each speaker's cloned voice in the target language, and remixes it over the original background audio. Per the vendor's API pricing page at time of writing, Dubbing v2 supports 92 languages, accepts MP3, MP4, WAV and MOV, and is priced at $2.20 per minute of media via API. The older v1 flow survives at $0.33 per minute watermarked ($0.50 without watermark). The claim worth testing is right in the feature description: that emotion, timing, tone and speaker identity survive translation. In my experience it largely does — and it is expensive.

What it actually does

Automatic Dubbing (Dubbing v2) is the one-call flow: upload media, pick target languages, get back dubbed output with the original speakers' voices preserved and the original soundtrack retained underneath. Dubbing Studio is the editor-grade workspace — available from the $6/month Starter plan upward — where you can work at the track level: adjust translations, re-generate individual segments, manage speakers, and control the mix. Both draw from ElevenLabs' shared credit pool on subscription plans (Free 10k credits/month, Starter 30k, Creator 121k, Pro 600k, Scale 1.8M, Business 6M at time of writing), while the API bills Dubbing v2 per minute in dollars. Concurrency is three simultaneous dubbing jobs on all self-serve plans (ten by default on Enterprise), applied per workspace and per model — v1 and v2 jobs count separately. Failed or cancelled jobs are refunded automatically, and free-tier dubs carry a watermark (Dubbing v2 has no watermark toggle; the discount-for-watermark option existed only in the legacy v1 flow and Dubbing Studio).

The surrounding product line is what makes the dubbing credible: the same voice cloning (instant cloning from short samples, professional cloning from longer recordings) and the same speech models feed the dubbing pipeline, so the output quality ceiling is the TTS ceiling. That is the real story here — ElevenLabs' dubbing quality is downstream of years of voice-model work, and it shows.

My testing notes

I ran a short multi-speaker video through the automatic pipeline, and two things stood out. First, speaker preservation: with distinct male and female voices in the source, the dubbed output kept them distinct and recognizably "the same people" in the target language, with pacing that tracked the original speech rather than reading like a parallel narration. Background music and ambience survived the remix without the pumping artifacts I've heard from dubbing tools that re-synthesize the whole track. Second, timing: lip-sync is not attempted (this is voice-and-timing dubbing, not lip-sync video), but segment timing stayed close enough to the original that off-camera and b-roll content felt natural. On-camera mouth movement obviously won't match — for talking-head content where faces are prominent, this is a real limitation versus lip-sync tools.

The credit economics are the part I'd model before committing. Third-party breakdowns put Dubbing v2 at roughly 13,500 credits per minute on the credit system, which means a $6 Starter plan covers only a couple of dubbed minutes before you're buying more — the widely cited API figure of $2.20 per minute is the honest unit cost, and it is roughly 4–7× the legacy v1 per-minute rates. For a creator dubbing a 10-minute video into three languages, that's around $66 of API-equivalent value; for a media company with a back catalog, the numbers escalate fast. My advice: use Dubbing Studio's editing on high-value content where you'll actually rework segments, and reserve v2 one-shot dubbing for content where a single pass is acceptable.

The workflow around the API is worth understanding before you commit. A dubbing job is submitted with source media and target languages, runs asynchronously through the pipeline, and returns dubbed tracks; the docs are explicit about failure semantics (automatic refunds, and guidance to cancel and resubmit jobs stuck in Queued or Loading without losing credits), which tells me the job queue has had its operational bumps and the vendor has chosen honest refund behavior over fine print. Model-specific terms for Dubbing v2 exist as a separate document, and free-tier dubs are watermarked automatically — there is no watermark-for-discount toggle on v2, a change from the legacy flow that the docs call out directly. For production use I would route everything through the API rather than the web app, both for repeatability and because the per-minute API pricing is clearer to forecast than credit consumption.

There is also an ethics and rights dimension I take seriously with voice cloning in dubbing. ElevenLabs gates professional voice cloning behind verification and consent processes, and dubbing that preserves a speaker voice in a language they never spoke is exactly the kind of capability that needs permission and provenance. For corporate and media customers that means thinking about releases and disclosure before dubbing real people; for content creators dubbing their own material, the consent question answers itself. The platform terms and the v2 model-specific terms place obligations on users here, and I would read them as part of the buying decision rather than after it.

Finally, comparison shopping. If your requirement is faithful voice identity across languages and a retained soundtrack, this is the strongest output I have tested and the premium is defensible. If your requirement is lip-synced video with matching mouth movement, this product is not that and cheaper lip-sync tools exist; if your requirement is bulk low-stakes dubbing at minimum cost, several competitors undercut the per-minute rate substantially. The right question is not whether $2.20 a minute is expensive in the abstract — it is what a viewer reaction to a flat synthetic voice versus their own voice is worth for the specific content you are shipping.

What it does not do

No lip-sync — mouths don't move with the translated audio, so this is not a replacement for dedicated lip-sync dubbing on face-forward video. No live or realtime dubbing (the docs confirm this). No free ride: the free tier's dubs are watermarked automatically, and the credits go quickly at dubbing rates. Speaker diarization and separation of overlapping speech can produce mis-assigned segments in dense multi-speaker audio, in my experience requiring Studio cleanup. The product is young — Dubbing v2 is newer than the TTS line, and the docs note differences in behavior between v1, v2 and Studio (refund behavior, watermarking) that suggest the surface is still settling. And on pure per-minute cost, competitors undercut it substantially; if budget is the binding constraint and voice fidelity is secondary, cheaper options exist.

Verdict

ElevenLabs Dubbing rated 4.1 on my rubric: the core functionality — voice-preserving translation with retained background audio across 92 languages — is the strongest dubbing output I have tested, and Studio's segment-level control is real production tooling. Value is the deduction: $2.20 per minute via API and steep credit consumption put it at the top of the price range, and the young product surface adds friction. Usability is good once you understand credits; ecosystem and reliability benefit from the mature TTS platform underneath. If dubbing quality determines whether your content gets watched in another language, this is the tool I'd pay the premium for. If you are bulk-dubbing hundreds of low-stakes clips, do the math first — the per-minute cost is the story.

Sources

• https://elevenlabs.io/pricing
• https://elevenlabs.io/pricing/api
• https://elevenlabs.io/docs/capabilities/dubbing
• https://elevenlabs.io/docs/eleven-creative/products/dubbing
• https://elevenlabs.io/dubbing-studio
• https://startpinch.com/guides/elevenlabs-dubbing-pricing
• https://voixie.com/voixie-vs-elevenlabs
• https://mydubly.com/guides/elevenlabs-dubbing-alternative
• https://owlwebmasters.com/en/services/elevenlabs
• https://elevenlabs.io/blog/introducing-dubbing-v2
• https://elevenlabs.io/blog/dubbing-api
• https://elevenlabs.io/dubbing-v2-model-specific-terms