Sonic is a blazing fast, lifelike generative voice API (🚀 135ms model latency). Build high quality, real time voice experiences with a diverse voice library, instant voice cloning, voice mixing, and voice design with speed and emotion control.

Hi all! I'm Brandon, cofounder of Cartesia! We’re thrilled to be sharing our work on Sonic, the fastest human-like voice API. As a company, we’re obsessed with making intelligence faster, cheaper, and more accessible for everyone. And we believe that we need fundamentally new AI architectures to get there. We founded Cartesia to build long-lived real-time intelligence for every device, starting with a kernel of architectural innovation: state space models. My cofounders Karan, Albert, and Chris pioneered State Space Models (SSMs) in academia, and our models like S4 and Mamba are now widely adopted across industry and academia. Sonic is built on a new state space model architecture we’ve developed for low latency modeling of high resolution signals - more technical details here soon! Thanks so much for checking us out! Product Hunt users who sign up for a paid tier get 10% off for their first 2 months through next Friday, June 7. Email support@cartesia.ai with the subject line Cartesia x Product Hunt Promo to claim.

@bclyang This is super impressive! It will unlock tremendous potential in building AI capabilities. Congrats!

@bclyang Super fun product! Is there any plan to improve the writer's content in addition to man/woman/fun-like voice? It fits so well with the current product (ie. reading what the writer writes), might as well improve it while you are at it ;) Really cool! I need an avatar soon too :D Keep pushing the boundary here!

@bclyang Hey Brandon - this is incredibly cool. Love the platform approach you and the team have taken to this API and the ability to have extremely low latency in creating interactive and conversational agents for Customer Service chatbots, voice agents, tutoring, or language services. Congrats on the launch - best of luck!

@bclyang impressive! The model performance, the platform approach, and ecosystem potential. Love to see more use cases landed with frictionless experience across voice and text, and future video too. Congrats for the launch!