About Applied Brain Research
Applied Brain Research develops low-power AI for speech and audio on small devices. Its TSP1 chip is designed to run speech recognition on the device itself instead of in the cloud, using state-space models, a type of neural network that processes sequences efficiently, along with the Legendre Memory Unit developed at the University of Waterloo. The company also maintains Nengo, an open source tool for building neural models. It is based in Waterloo, Ontario, and grew out of Chris Eliasmith's research lab.
Applied Brain Research was founded in 2013 and is based in Waterloo, Canada. It is listed under Neuromorphic Computing. Reported funding is Oversubscribed seed round (Jan 2026, led by Two Small Fish Ventures).
Compute Technologies
- Neuromorphic
What Applied Brain Research works on
Neuromorphic Computing
Neuromorphic computing builds chips that copy some features of the brain. Instead of a central processor and separate memory, a neuromorphic chip has many small units that act like neurons, each with its own local memory, joined by connections that act like synapses.
