TEAM

PhDs, professors and researchers

From the world's top universities in audio-visual human research.

▚ THE TEAM ▞

We exist to break every barrier to understanding.

An Zhu Liu

Jibin Song

Mingi Kwon

Jaeseok Jeong

Prof. Xu Zheng

Taehwan Lee

Tonghyun Lee

Michael Chan

Prof. Youngjung Uh

▚ RESEARCH ▞

We're building world translation: not just voice and lips -- the scene, the sound, the human.

Joint audio-visual generation

The voice and the face generated by one model, so identity can never split: the field we started with JAM-Flow.

Real-time human animation

The most realistic real-time humans today: SoulX-LiveAct generates a person at 20 frames per second, stable for an hour.

Fast video generation

Distillation, guidance, and sampling: the research that makes diffusion models fast enough for real time.

Editing inside diffusion models

Editing inside frozen diffusion models, from finding h-space to controlling style and removing on-screen text.

Multimodal perception

Reading the scene before editing it: who is talking, from visual and audio together, in any environment.

Since 2021, over 10,000 citations and 100+ papers at ICLR, NeurIPS, CVPR, ECCV, and more.

▚ TESTED ON THE SAME CLIPS ▞

Sounds +27.8% more like you

ElevenLabs: +270% more background-sound error; the laughter, the music, the ambience

Makes 54.7% fewer translation errors; changes or loses important ideas

Sounds +182.9% more like you

YouTube: +232.6% more background-sound error

Makes 70.3% fewer translation errors; changes or loses important ideas

100.3% of expert human translation quality

5× cheaper than human translators

vs $22–45 per spoken minute, translation alone