Cartesia is an artificial intelligence company focused on building real-time, multimodal intelligence that can operate across diverse environments. Its work centres on voice AI, developing models that enable more natural and expressive interactions between humans and machines. The company's technical foundations lie in State Space Models (SSMs) and large-scale foundation model research.
The firm's product line includes Sonic, a text-to-speech model designed to produce human-like vocal expression, and Ink, a speech-to-text model engineered for real-time voice applications. Both are aimed at developers building voice-driven products.
The founding team met as PhDs researchers at Stanford's AI Lab, where they helped pioneer State Space Models. The company applies a design-minded approach to its engineering, with the stated goal of creating tools that help developers build voice applications that connect with users.
