Welcome to AI with Sally, a podcast that takes a closer look at some of the most interesting technology stories on artificial intelligence and machine learning. We’ll hear about the latest in hardware and software that has a big impact on the world of AI. I’m your host, Sally Ward-Foxton.
In this podcast episode, Sally speaks to SynSense’s Dylan Muir about the company’s design win for its neuromorphic processor in a toy robot that recognizes gestures, the company’s two processors and the differences between them, as well as the synergies between dynamic vision sensor cameras and neuromorphic processors.
Sally Ward-Foxton
Neuromorphic processor company Synsense has two families of spiking neural network accelerator – Dynap-CNN is for accelerating the spiking equivalent of convolutional networks, used in image processing, and recently introduced Xylo is for ultra-low power processing of time-series sensor data like audio or vibration.
I last spoke to SynSense about a year ago when the company partnered with dynamic vision sensor company Prophesee. At that time, SynSense had already integrated its Dynap-CNN processor with a dynamic vision sensor – or DVS sensor – from another company called IniVation, in a module known as Speck. DVS sensors, as you’ll recall, produce an output only when pixel data changes. This means you get a very low-bandwidth version of the scene, only showing moving objects. The future Prophesee module will be higher resolution than the existing IniVation Speck module, for different use cases.