28 July, 2018
Aimed at professional engineers, the AIY Edge TPU Dev Board and AIY Edge TPU Accelerator are powered by Google’s Edge Tensor Processing Units (TPUs)—specialized chips that facilitate AI assignments.
Two years, ago, Google introduced its cloud TPU (a third version of which is coming soon), optimized for training machine-learning models. The tech titan’s latest additions will run inference, i.e., actually carry out tasks like speech or image recognition.
Considered a speedier, more efficient way of interpreting data, this sort of on-device machine learning was designed for enterprise jobs. So don’t expect to find these chips embedded in your next smartphone.
“To jump-start development and testing with the Edge TPU, we’ve built a development kit,” Injong Rhee, Google Cloud’s vice president of IoT, explained in a blog announcement.
That includes the AIY dev board and accelerator, both set to hit digital shelves in the US this October, before rolling out to more countries. Engineers can request early access online.
The all-in-one AIY dev board—a Raspberry Pi-esque baseboard with all the peripheral connections you need for prototyping—features a removable system-on-module (SOM) daughter board for future project scaling.
“On-device ML is still in its early days, and we’re excited to see how these two products can be applied to solve real-world problems,” AIY project director Billy Rutledge wrote in a separate blog.
“This lets you build and train ML models in the cloud, then run those models on the Cloud IoT Edge device through the power of the Edge TPU hardware accelerator,” Rhee boasted.