SynSense

SynSense is a neuromorphic engineering company specializing in ultra-low-power sensory processing hardware and application solutions. Founded in March 2017 in Zurich, Switzerland, the company moved its headquarters to Chengdu, China in 2020 while maintaining operations in Zurich. SynSense develops neuromorphic technology algorithms and hardware designs, focusing on IoT real-time signal processing and AI edge computing. The company's core product is the Speck dynamic visual intelligence System-on-Chip (SoC), which combines a low-power Spiking Neural Network (SNN) vision processor with an event-based sensor. Speck utilizes full asynchronous circuit design to perform both neuromorphic sensing and computing, capturing real-time visual information, recognizing objects, and performing vision-based detection and interaction functions. The chip operates on less than one milliwatt of power consumption with 5-10 millisecond end-to-end latency. SynSense's technology leverages event-based vision, which captures changes in a scene rather than recording entire scenes at a static frame rate, reducing data redundancy and latency while protecting privacy. In February 2024, SynSense acquired iniVation, a developer of neuromorphic vision systems, to strengthen its position in the neuromorphic technology market.

Key customers and partnerships

SynSense has engaged in technological exploration with BMW to advance the integration of neuromorphic chips in smart cockpits. In April 2023, the company unveiled the world's first neuromorphic programmable robot at the China Shantou International Toy Fair in collaboration with QunYu, a leading Chinese company for programmable intelligent toys. The robot, which incorporates SynSense's Speck chip, can recognize eight different human body postures and mimic human movements. SynSense and QunYu have signed a strategic cooperation agreement to jointly launch neuromorphic interactive blocks. Additionally, in November 2023, SynSense announced a collaboration with the University of Groningen's CogniGron to evaluate its low-power neuromorphic sensory processors and develop applications for its neuromorphic hardware.