With the explosion of machine learning deployments on the network edge over the past few years, there’s a growing demand for performant and efficient hardware, purpose built for the unique compute demands of neural networks and ML algorithms.
To answer this demand and bring AI acceleration to Edge Impulse developers, we are excited to announce our partnership with Alif Semiconductor. Alif has developed the world’s first implementation of the Arm Ethos-U55 micro neural processing unit (microNPU) and Cortex-M55 low-power MCU in their Ensemble and Crescendo device families.
These devices support continuous monitoring of the environment using sound, vibration, or images on an extremely low-power energy budget. Then, when a trigger event is detected, high-performance speed neural compute systems churn through deep learning workloads hundreds of times faster than prior-generation embedded processors.
Alif has opened up an early access beta program for those interested in developing new applications made possible by their innovative hardware, and we’ve likewise added early access support for the Ethos-U55 in Edge Impulse! If you are interested in joining the beta program and would like to learn more, reach out to firstname.lastname@example.org or edgeimpulse.com/contact.
We’ll be announcing more exciting news with Alif in the near future, so stay tuned!