Introducing EVI 3: general voice intelligence

Подробнее о видео

EVI 3 is a speech-language model that can understand and generate any human voice—not just a handful of speakers. With this more general voice understanding comes a richer understanding of tune, rhythm, timbre, and speaking style, and greater realism and expressiveness.

EVI 3 sits at the frontier of low-latency intelligence, thanks to our latest voice-to-voice architecture.

EVI 3 can instantly generate a new voice and personality from any prompt in under a second. You can speak to a “raspy Australian history buff,” “calming British matriarchs,” “excited Caribbean musician,” — any voice you can think of.

EVI 3 transforms its delivery implicitly or on command—stammering anxiously, debating enthusiastically, or whispering intimately. We’re still discovering new capabilities.

You can try EVI 3 today and create a personal voice: demo.hume.ai