Frontier models
We train audio intelligence systems that understand motion, timing, and texture.
About Us
The frontier AI research lab building audio models for videos.
Sound has always been the hardest part of the visual stack to get right — until now. We build state-of-the-art generative models that transform how the world is heard, from individual creators to the world’s largest production pipelines.
From our lab in Berlin, a small but exceptional team is creating what comes next.
Creative workflows
Our tools help creators move from raw picture to finished sound faster.
Production scale
Mirelo is built for individual makers and the largest visual production pipelines.
Backed by leaders in AI and technology
Andreessen HorowitzAtlanticEllipsisVELATriplePoint CapitalIndex Ventures
Jakob Uszkoreit
Léon Bottou
Bernhard Schölkopf
Yariv Adan
- Jakob Uszkoreit
- Léon Bottou
- Bernhard Schölkopf
- Yariv Adan
- Jürgen Galler
- Thomas Hofman
- Matthias Dantone
- Roméo Walter
- François Charton
- Pierre-Henri Gabriel
- Thomas Wolff
- Arthur Mensch
- Robin Rombach
- Burkay Gur
- Coco Mao
- Alex Berriche
- Laura Modiano
- Antoine Le Nel