Cartesia

Cartesia builds real-time multimodal AI foundation models and developer tools, with a focus on voice intelligence. The company's founding team met as PhDs at Stanford's AI Lab, where they pioneered State Space Models (SSMs), an approach to efficient large-scale foundation models that underpins Cartesia's technical work. The team combines expertise in model innovation with systems engineering, and has deliberately moved beyond academic research to ship products used by developers.

Two models form the current product line. Sonic is the flagship text-to-speech model, designed to produce speech that laughs, emotes and sounds natural. Ink is a speech-to-text model built specifically for real-time voice AI. Together they target voice applications and developer tools, where latency and expressiveness are the defining constraints.

Technical work spans State Space Models, efficient large-scale foundation models, real-time multimodal intelligence, and the systems engineering required to run them. The company describes its approach as design-minded, aiming to build cutting-edge experiences that connect with people, and its stated focus is on helping developers build voice applications.