Cartesia Secures $27M to Expand Real-Time AI with State Space Models and Sonic Voice Technology

Cartesia, a startup specializing in state space models (SSMs), has raised $27 million in funding, led by Index Ventures, to further develop its innovative AI technology. SSMs, a newer AI architecture, offer significant advantages over traditional transformer-based models, including higher efficiency, lower latency, and the ability to run on devices without reliance on data centers. Cartesia’s flagship product, Sonic, is a low-latency voice model capable of generating lifelike speech with less than 90 milliseconds of latency. Sonic already serves hundreds of customers through its API, showcasing its potential for real-time applications. The funding will enable Cartesia to expand its real-time AI applications across various industries, including customer service, robotics, and gaming. The company’s long-term goal is to develop multimodal AI models capable of processing diverse inputs such as text, audio, and video, aiming to make real-time intelligence accessible on any device. Cartesia was co-founded by Stanford researchers Karan Goel, Albert Gu, Arjun Desai, and Brandon Yang, who have pioneered SSMs as a more efficient alternative to transformers. The funding round also included participation from A* Capital, Conviction, General Catalyst, Lightspeed, SV Angel, and several prominent angel investors.