ElevenLabs Ends Beta and Launches AI Voice Model 'Eleven Multilingual v2' Supporting 30 Languages

Voice AI platform ElevenLabs has officially ended its beta phase and launched Eleven Multilingual v2, a new foundational deep learning model that supports about 30 languages. With this update, media companies, game developers, and content creators can dramatically improve content accessibility across a wide range of languages worldwide.

Eleven Multilingual v2 automatically identifies about 30 languages from input text to generate highly natural, 'emotionally rich' audio. In particular, when using synthetic or cloned voices, the key feature is that the speaker's unique vocal characteristics and original accent are preserved across all languages.

Key updates include the following:

• Expanded multilingual support: Support for about 30 languages, including Korean, Chinese, and Japanese
• Consistent voice preservation: The speaker's unique tone and accent are preserved even when the language changes
• Professional Voice Cloning integration: A user's voice can be perfectly cloned and reproduced in multiple languages

ElevenLabs plans to introduce collaboration mechanisms in the future that will allow users to share voices within the platform and participate in developing new audio.