ElevenLabs Introduces "Dubbing v2" — Next-Generation AI Dubbing That Preserves Emotion and Pacing
ElevenLabs unveiled a new dubbing model, “Dubbing v2.” Its standout feature is the ability to translate naturally across more than 90 languages and accents while preserving the emotion, performance, and intent of the original content.
Details
- Language and accent coverage: Supports over 90 languages and accents
- Preserving emotion: Addresses the common shortcoming of past AI dubbing — flat dubs detached from the original context — by preserving emotion, pacing, and pauses
- How it’s delivered: Available through Dubbing Studio, built on ElevenLabs’ existing speech synthesis and multilingual model technology
- Intended use cases: Aimed at creators and businesses that need to localize video content, podcasts, and similar media into multiple languages
How to try it
- Available through Dubbing Studio at elevenlabs.io
- Existing ElevenLabs account holders can simply upload a video or audio file and select a target dubbing language to get started
- Pricing follows ElevenLabs’ existing subscription plans