Organisation · company · US

Cartesia

Low-latency, real-time voice generation built on state-space model architectures.

Reference profile

Cartesia is a US voice-AI company, founded in 2023 by researchers from Stanford's AI lab, including Karan Goel and Albert Gu. It builds low-latency speech-generation models — its Sonic line — around state-space model architectures, an alternative to the transformer designs used in most large models that its founders argue is better suited to real-time, streaming audio. The company reported raising a $64m Series A in 2024 led by Kleiner Perkins, and a later $100m round with backers including Nvidia.

Category
Generative media
Founded
2023
HQ
San Francisco, US
Funding
$100m round (backers include Nvidia and Kleiner Perkins); $64m Series A (2024)