Artificial intelligence (AI) audio corporations ElevenLabs unveiled a new AI dubbing model, "Dubbing v2," on the 16th.
The model is distinguished by improvements to the loss of emotion and delivery that had been cited as a limitation of existing AI dubbing. It analyzes the speaker's actual expressive style—such as the emotion, tone, intonation, and pacing of the original voice—and naturally reproduces it in more than 90 languages.
It also supports multilingual dubbing that goes beyond simple translation to maintain the immersion and character of the original content through automatic voice cloning, automatic voice timing adjustment, and context-based translation.
The company expects "Dubbing v2" to support the global expansion of Korean drama, films, webtoons, games, and YouTube content. A ElevenLabs official said, "Korea holds powerful content assets that captivate viewers worldwide, but previously, translation, scriptwriting, voice acting, audio editing, and timing adjustments required massive production steps and expense for global expansion," adding that "Dubbing v2" will help shorten this work time.
Hong Sang-won, head of ElevenLabs Korea, said, "Making translated voices feel as if the original speaker actually spoke that language has been one of the biggest challenges in AI dubbing," adding, "The newly introduced 'Dubbing v2,' based on precise emotional reproduction and excellent sync, will be a powerful bridge for domestic creators, marketers, and content corporations to communicate more deeply and naturally with people around the world beyond language barriers."