Introducing Turbo v2.5, low latency text to speech in 32 languages

ElevenLabs has introduced Turbo 2.5, a new model that enhances text-to-speech capabilities by making them three times faster in 32 languages, including Hindi, French, Spanish, and Mandarin, reaching nearly 80% of the global population. The update also marks the first-time support for Vietnamese, Hungarian, and Norwegian, while improving English performance by 25%. Users interested in utilizing these advancements can begin building with ElevenLabs' API, and those needing volume discounts and higher concurrency can explore enterprise options. The company recently demonstrated its multilingual AI voice technology with NVIDIA ACE at Computex, where NVIDIA's founder and CEO, Jensen Huang, delivered parts of his keynote in both English and Mandarin using ElevenLabs' technology.