Familiar
New accelerator entry · 2026-08-12 · the record
In their words
Audio-visual translation infrastructure for all video, live and recorded -- the first and only that works in real time. World translation: not just changing voice and lips into a new language, but understanding the scene, the sound, and preserving the human. 🦋 Movies & TV, Micro-dramas, Live Broadcasts, Live Shopping, Corporate & Education, Ads, Creators & Podcasts, Platforms. Day-and-date, in every territory. Founders An Zhu Liu, Jibin Song, Mingi Kwon, and Xu Zheng lead a team of PhDs, professors, and researchers from the world's top universities who started the field of joint audio-visual human models and co-authored the current state of the art in real-time human animation. Since 2021: over 10,000 citations and 100+ papers at ICLR, NeurIPS, CVPR, ECCV, and more. Translation needs 3D world understanding, permanence of identities. A translator has to know who's speaking: where every sound in the dense polyphony originates (laughter, layered crowds). Laughter here, a short yell there, a voice from off-screen, someone on screen mouthing words before being interrupted. Everyone else scaffolds separate models and pre-processing; a unified model simulates the scene. It's a spatio-temporal reasoner: predicting continuity across frames forces it to internalize 3D geometry, spatial reasoning, and permanence -- of objects, people, and the environment itself. The scene state survives the edit: when the shot pans or cuts away, someone off-screen still exists, still owns their voice, and is still there when the camera comes back. Multiple people, side views, off angles, dense polyphony, fast motion -- the hard cases hold. Vs ElevenLabs: they lose the laughter, the music, the ambiance -- 270% more background-sound error -- and ours sounds 27.8% more like you. That means no recasting, no ADR, no M&E stems required -- it works from the final mix. Vs human translators: 100.3% of their quality -- scalable, real time. Published benchmarks: https://thefamiliarlab.com/benchmark
What the record says
Founders, titles and bios as the company published them. Photographs are our own copies of the ones on that record, so they do not break when the original link expires.
Against the 209 companies on file from Summer 2026. Median team 2, largest 30.
The nearest companies on file from the same batch, side by side with this one.
| Company | Batch | Sectors | Listed |
|---|---|---|---|
| Frontier Computing | Summer 2026 | Artificial Intelligence, Hard Tech, Hardware | 2026-08-31 |
| Ethos Space Resources | Summer 2026 | Hard Tech, Exascale Computing, Energy | 2026-08-23 |
| Spectre Intelligence | Summer 2026 | Biotech, Investing, Trading | 2026-08-22 |
| Edgerun | Summer 2026 | Robotics, Defense | 2026-08-21 |
| Allia Health | Summer 2026 | Artificial Intelligence, Healthcare, Mental Health | 2026-08-19 |
| Familiar (this page) | Summer 2026 | Generative AI, B2B, Entertainment | 2026-08-12 |
Click a column heading to sort. Comparison is against records held here, not against every company that exists.
The full profile
The rest of this profile is open on any paid or trial plan
It costs no credits. It is a plan check and nothing else.
- Source Check — verify An Zhu Liu's other claims against the registers that hold them.
- INVEST — compare this company against the peers above, on the numbers.
- RECRUIT — see who is already on the team.
- Press Requests — connect with An Zhu Liu for a story.
You land back on this company when you are in.
Not ready to upgrade? Keep this one in view instead.
Watchlist FamiliarYou get an alert in MentionFox the next time this company shows up.
Is this you?
If you're behind Familiar: acceptance into a YC batch is one of the few weeks a young company gets looked up by people who matter — investors, other founders, press. FoxPlug turns the commits, deploys, and releases you're already making into that story automatically, so anyone who looks finds something current. FoxChat puts a guide on your own site so the people YC's own page just sent you don't read one line and leave.
Sourced from the company's Y Combinator directory entry. Figures are read directly from the source record; names have not been matched against other people.