TEAM
PhDs, professors and researchers
From the world's top universities in audio-visual human research.
Stanford
Yonsei
HKUST
We exist to break every barrier to understanding.
Imagine a world where everything we watch plays seamlessly in every language. English speakers get the best of Korea, China, and everywhere else. Everyone else gets English content without subtitles.
Many of us on the team learned English from television. Our CEO, Andrew, arrived in Canada as a kid without the language: the only Chinese kid in his class, unable to join in, until the shows he watched every night taught him the words. The language barrier has been a sore spot for us ever since.
We're building multimodal translation: not just voice and lips -- the scene, the sound, the human.
The voice and the face generated by one model, so identity can never split: the field we started with JAM-Flow.
The most realistic real-time humans today: SoulX-LiveAct generates a person at 20 frames per second, stable for an hour.
Distillation, guidance, and sampling: the research that makes diffusion models fast enough for real time.
Editing inside frozen diffusion models, from finding h-space to controlling style and removing on-screen text.
Reading the scene before editing it: who is talking, from visual and audio together, in any environment.
▚ TESTED ON THE SAME CLIPS ▞