Berlin, Germany - September 20, 2023 - Advancing the mission of democratizing artificial intelligence, Coqui, a pioneering AI startup established by industry visionaries from Mozilla, proudly introduces the open-access release of their trailblazing foundation model for generative voice AI: XTTS. Building on the success of Coqui’s previous code and model releases, which garnered millions of downloads, XTTS promises even broader impacts.
Enhancing its global appeal, XTTS supports speech in an expanded set of 13 languages, leveraging advanced generative AI techniques to produce faster and more refined speech, now fully integrated with TTS. Coqui has already achieved being the #1 trending repo on GitHub, the #1 trending space of the week on Hugging Face, and #1 trending on Replicate.
XTTS's groundbreaking features include:
- Voice cloning from just a 3-second audio clip.
- Emotion and style transfer during cloning.
- Cross-language voice cloning capabilities.
- Multi-lingual speech generation.
- A superior 24khz sampling rate.
At present, XTTS-v1 supports English, Spanish, French, German, Italian, Brazilian Portuguese, Polish, Turkish, Russian, Dutch, Czech, Arabic, and Mandarin Chinese.
Hugging Face, a renowned platform in the AI community will host this transformative model, underscoring the profound impact of this release.
Berlin's voice clone AI startup, Coqui, recently celebrated a successful $3.3 million seed funding round.
Joshua Meyer, Coqui’s Co-Founder, shared, "Our team is deeply rooted in open science. Originating from the Mozilla machine learning group, we championed breakthroughs in open code, models, and data. At Coqui, our vision is clear: to construct foundation models for generative Voice AI that will soon become omnipresent. This isn’t our first collaboration with Hugging Face, and we eagerly anticipate future foundation model launches on their esteemed platform!"
Julien Chaumond, Hugging Face CTO and Co-Founder, expressed his enthusiasm:
"Huggingface loves collaborating with open-source AI leaders like Coqui. The XTTS release in particular was extremely successful and gathered lots of positive engagement from the community. Generative AI is making waves and should be openly accessible to everyone! We look forward to more great collaborations with Coqui!"
Setting a New Standard in Open Model Licensing:
Leading the charge in open innovation, Coqui, in partnership with esteemed open-source licensing expert Heather Meeker, has crafted the Coqui Public Model License (CPML). CPML stands as a testament to Coqui's dedication to transparency, collaboration, and open AI advancement.
Hugging Face Collaboration: Symbolizing the unity of the open research domain, Coqui and Hugging Face's partnership aims to make XTTS universally accessible. Hugging Face's pivotal role in hosting XTTS ensures an intuitive user experience, catering to developers, researchers, and creative entities globally.
The XTTS model's launch marks a monumental shift in voice AI models, bridging the longstanding divide between open-access and proprietary systems. This pivotal move heralds a transformative era in globally accessible, superior voice AI technology.
XTTS: A New Paradigm in Voice AI:
Unparalleled Quality: XTTS is setting unparalleled standards in voice AI with its unmatched voice generation quality.
For those inspired by XTTS's capabilities, Coqui unveils two enticing offerings:
Coqui and Hugging Face are pioneering a new chapter in the generative AI saga. As Joshua Meyer aptly puts it, "XTTS stands tall in the realm of voice, akin to Stable Diffusion for images and Llama for text. Coqui spearheaded the research and development that made XTTS possible, and Hugging Face amplifies its reach to the broader community."
Coqui warmly invites global innovators to explore and redefine the frontiers of generative voice AI with XTTS, fostering a world where revolutionary technology is universally accessible and catalyzing humanity's collective evolution.
Demo: https://huggingface.co/spaces/coqui/xtts
Docs: https://tts.readthedocs.io/en/latest/models/xtts.html
For media inquiries and more details, reach out to:
Cigdem Oztabak I PR Representative at cigdem@pirix.co
About Coqui:
Coqui is a forward-thinking machine learning company with a strong foundation in open-source principles. Founded by industry leaders with a rich history of open-source contributions, Coqui is committed to democratizing AI technology and making cutting-edge solutions accessible to all. With the release of the XTTS model and the introduction of the Coqui Studio, API service, and CPML, Coqui continues to pave the way for a more inclusive and innovative AI landscape.
Media Resources:
- Coqui Homepage: https://coqui.ai
- Coqui Studio: https://coqui.ai
- Coqui API: https://docs.coqui.ai
- Github Repository: https://github.com/coqui-ai/tts
Media Contact
Company Name: Coqui
Contact Person: Cigdem Oztabak
Email: Send Email
City: Berlin
Country: Germany
Website: coqui.ai
