Elevenlabs specializes in generating high-quality synthetic voices. Now the startup presents its latest model.
The new model "Eleven Multilingual v2" supports 28 languages, which it automatically recognizes and converts from text to speech. According to Elevenlabs, it achieves even better voice authenticity than its predecessor.
The characteristics of a speaker's voice, whether synthetically generated or a voice clone, are preserved in all languages, so one voice can be used for all languages.
The first version of Elevenlabs already featured some of the highest quality voice generations on the market. A demo of the new version can be heard in the video below.
Elevenlabs now supports Chinese, Korean, Dutch, Turkish, Swedish, Indonesian, Filipino, Japanese, Ukrainian, Greek, Czech, Finnish, Romanian, Danish, Bulgarian, Malay, Slovak, Croatian, High Arabic, and Tamil. English, Polish, German, Spanish, French, Italian, Hindi and Portuguese were and are supported.
Elevenlabs plans to launch an AI voice sharing platform
With the release of v2, Elevenlabs also ends its beta phase. According to the company, its technology now has more than one million users worldwide. In the future, a new feature will allow users to share voices on the platform. The ambitious vision is to make all content available in any language and in any voice.
The target audience is media companies, game developers, publishers, and authors. Customers already using the technology include D-ID, Storytel, ScienceCast, TheSoul Publishing, Embark Studios, Paradox Interactive, and MNTN. The company's proprietary audio technology was developed entirely in-house, according to Elevenlabs.