There's a lot happening around generative AI in text generation (GPT 3, Bloom), image generation (Stable diffusion, OpenAI), video generation (Makeavideo, runwayML). Today I'm excited to share our very first Large Language Generative AI model for speech generation, called Peregrine. Peregrine is a text to speech model that is capable of generating truly expressive human-like speech including emotions and laughter! It's also capable of cloning a voice with near perfection results. Happy to hear what the community thinks of it!

Wow! Congrats on the launch 🚀 will download and try it now

Congrats on the launch. The voices sound great!

Hello, I'm a bit disappointed by your move! When I bought your LTD, you told us that all futures voices would be included and now you create "ultra realistic voices", not included in the LTD... Soon, you'll launch a new LTD which will include these Ultra realistic voices, then you'll create mega ultra realistic voices... It's not fair.

this is mind blowing stuff, so far the reason i did not used text to speech was robotic voice you guys seems to have solved, going to add it to my blogs now

Voices are awesome. Can we upload our own voices also here? Thanks

This is the best place for a proper voice over, hands down. Thanks guys, keep up the good work!