Dubbing v2
Dubbing v2 conditions the translated audio on the original performance. This differs from a conventional recognition, translation and cloned-speech pipeline. It aims to carry emotion and timing into the target language.
Preview
Performance and output
Cloning strength trades resemblance to the source speaker against natural delivery in the new language. Higher settings can retain more of the original accent. API transcript editing and regeneration require an Enterprise plan.ElevenLabs Dubbing capabilities and alpha status
The video download is a preview and can differ from the source resolution. Audio output is at most stereo. Keep the original video master and use the dubbed audio when the original video quality must be retained.ElevenLabs Dubbing capabilities and alpha status
Practical guidance. Review translation, delivery and speaker identity separately. A familiar voice can still speak an incorrect translation.ElevenLabs: Dubbing v2 release
Specifications & access
| Detail | Value and conditions |
|---|---|
| Status | Preview |
| Launch date | |
| Access | API, Provider product |
| Model identifiers | Dubbing v2 Alpha |
| Current stage | AlphaElevenLabs Dubbing capabilities and alpha status |
| API launch | 6 August 2026, after the May product launchElevenLabs: Dubbing v2 API launch |
Related models
- Eleven v4
ElevenLabs · Eleven speech
Speech model for expressive narration, consistent character voices and multi-speaker dialogue.
Launch date API · Provider product - Multilingual Voice Changer v2
ElevenLabs · Eleven Voice Changer
Voice-conversion model that transfers a recorded performance to another voice across 29 languages.
Launch date not confirmedAPI · Provider product