Releases Eleven Multilingual v2

• ElevenLabs' Voice AI platform makes a significant stride towards dismantling language obstacles in content. Introducing the innovative deep learning model, Eleven Multilingual v2, it now boasts multilingual support across 28 languages.
• This development empowers media firms, game developers, publishers, and independent creators globally to significantly enhance the reach and accessibility of their content.
• Following a series of feature updates and enhancements since its inception in January, these new capabilities also signify the official conclusion of the company's Beta stage.
• ElevenLabs aims to ensure universal accessibility of all content, regardless of language or voice.

Today, ElevenLabs, a global frontrunner in voice AI technology, unveiled a state-of-the-art multilingual voice generation model. This innovative model can proficiently craft 'emotionally resonant' AI audio across close to 30 languages.

Built entirely on proprietary research, this breakthrough empowers creators to craft localized audio content tailored for diverse markets spanning Europe, Asia, and the Middle East. Over the past 18 months, ElevenLabs has delved deep into the nuances of human speech, innovating ways to grasp context and infuse emotion into speech synthesis, while also creating distinctive, novel voices.

Utilizing Eleven Multilingual v2, as text is fed into the ElevenLabs text-to-speech system, the advanced model seamlessly recognizes close to 30 written languages, producing speech with unparalleled authenticity.

Simultaneously, whether utilizing a synthetic or cloned voice, the distinct vocal traits of the speaker are consistently preserved across all languages, inclusive of their inherent accent. This allows the same voice to enliven content in 28 different languages.

This release comes after the introduction of Professional Voice Cloning for all creators on the platform. Accompanied by enhanced safety and security measures, this update enables users to craft an impeccable digital replica of their own voice, mirroring the original almost flawlessly. With today's update, your voice can now communicate in nearly 30 languages supported by the multilingual model.

Now supported languages encompass: Chinese, Korean, Dutch, Turkish, Swedish, Indonesian, Filipino, Japanese, Ukrainian, Greek, Czech, Finnish, Romanian, Danish, Bulgarian, Malay, Slovak, Croatian, Classic Arabic and Tamil.

They join previously available languages including English, Polish, German, Spanish, French, Italian, Hindi and Portuguese.

Building on its recent feature rollouts and continuous platform enhancements, ElevenLabs announced today its official exit from Beta. This significant milestone underscores the company's unwavering commitment to delivering state-of-the-art tools for its expansive global user base of over a million individuals.

Moving forward, ElevenLabs aims to implement a feature enabling users to share voices on the platform, paving the way for innovative audio development and nurturing the potential for human-AI synergies.

Mati Staniszewski, CEO and co-founder of ElevenLabs, comments:

ElevenLabs embarked on a journey with a vision: to render all content universally accessible across every language and voice. With the launch of Eleven Multilingual v2, we've edged closer to transforming that vision into reality, introducing AI voices that replicate the nuances of every dialect.

Our cutting-edge text-to-speech tools democratize audio quality, now empowering creators to broadcast in almost 30 different languages. As we move ahead, we're enthusiastic about expanding our language repertoire and voice variations through AI, tearing down the linguistic barriers that stand in the way of content accessibility. We at ElevenLabs hold a strong belief that these advancements in accessibility will be catalysts for heightened creativity, innovation, and diversity.

By streamlining the production of high-caliber multilingual audio content, ElevenLabs is positioning both enterprises and individual creators to craft content that strikes a chord across diverse cultural and linguistic landscapes.

For indie game designers and publishers, our multilingual audio tool unlocks avenues to tailor game narratives for global audiences, enabling them to engage users through authentic language experiences without compromising the integrity of the audio.

Educational institutions too stand to benefit, now equipped to offer learners accurate audio translations instantly, which not only enriches language understanding and pronunciation but also accommodates varied instructional methods and learning requirements of international students.

Moreover, creators can harness the power of ElevenLabs to amplify accessibility for those visually challenged or with distinct learning requirements by complementing visual media with multilingual speech.

Since its inception in January 2023, ElevenLabs has showcased a suite of AI-driven voice tools, from converting text to speech using synthetic voices to crafting a digital twin of one's own voice. The introduction of the multilingual speech synthesis tool is a testament to our unwavering commitment to global content accessibility.

Various sectors and creative industries have already embraced this innovative technology. It's been instrumental in helping indie writers produce audiobooks, giving voice to secondary characters in video games, assisting the visually impaired in accessing online materials, and even driving the world's premier AI radio channel. Furthermore, our collaborations span renowned content creators and studios, such as AI video creators D-ID, audiobook giant Storytel, scientific video platform ScienceCast, global content powerhouse TheSoul Publishing, stellar game developers like Embark Studios and Paradox Interactive, and the renowned media platform MNTN.