ElevenLabs Released New Speech AI Models
The company launched its v4 and v4 Turbo speech models with support for over 90 languages.
Updated on Sept. 28, 2026 in Artificial Intelligence
Live Poll
Do you trust the rapid advancement of voice AI technology in your daily life?
ElevenLabs has officially released its v4 and v4 Turbo speech models, designed for developers building voice agents and audio applications. These new models allow for voice cloning from just 10 seconds of audio and expanded expression controls.
Why it matters
By providing real-time audio generation that adjusts to textual context, these models aim to enhance the quality and versatility of AI-driven voice applications. The expansion into over 90 languages signals a push to scale these capabilities across global markets including Europe, India, and Brazil.
The v4 and v4 Turbo models support over 90 languages and require only 10 seconds of audio for voice cloning. Users can now stack multiple inline expression tags to adjust delivery based on the provided text.
The players
ElevenLabs
An AI company that provides speech synthesis technology and currently holds an $11 billion valuation.
Sequoia
A prominent venture capital firm that led the company's recent $500 million funding round.
The details
The models generate audio simultaneously as the language model creates a response, allowing for seamless integration in real-time voice agents. The update also features enhanced expression controls to provide more natural sounding speech for diverse use cases.
Timeline
September 28, 2026: ElevenLabs launched its v4 and v4 Turbo speech models.
The Tech Race
The AI sector is seeing a rapid shift toward enterprise-grade tools, with major companies now driving the majority of growth for developers. This launch positions ElevenLabs to compete in the high-stakes battle for voice-agent dominance against incumbent software providers.
Users can expect more natural-sounding voice interactions in applications that utilize ElevenLabs technology. Developers may also see improved workflow efficiency thanks to the new inline expression tags and reduced audio requirements for voice cloning.
The takeaway
The move to support over 90 languages makes AI voice tools accessible to a much broader global demographic. Businesses can now leverage these high-fidelity models to create more localized and expressive customer experiences.
Further reading
For more on the evolution of voice synthesis technology, explore the Artificial Intelligence section.
Live Poll
Do you trust the rapid advancement of voice AI technology in your daily life?
