Microsoft launches Azure Percept, its new hardware and software platform to bring AI to the edge

Microsoft launches Azure Percept, its new hardware and software platform to bring AI to the edge

Microsoft today announced its new hardware and software platform Azure Percept to bring more of its Azure AI services to the forefront. Percept integrates cloud tools with Microsoft’s Azure to manage devices with hardware and create AI models from Microsoft’s device partners.

The general idea here is to create an AI for businesses that provide end-to-end solutions for things like object detection, extraordinary identification, shelf analytics, end keyword spotting, and make it easier to implement from deploying to compatible hardware. To get it started, Microsoft today launched a hardware development kit with an intelligent camera for vision use (dubbed Azure Percept Vision). The kit features hardware-enabled AI modules for running models on the edge, but it can integrated with the cloud. Users will also be able to test their proof-of-concept concept in the real world because the development kit adheres to the widely used 80/20 T-slot framing architecture.

In addition to Percept Vision, Azure is also introducing Percept Audio for audio-centric use. “We’ve started with the two most common AI workloads of pressure, vision and voice, vision and sound, and we’ve given that blueprint so that manufacturers can take the basics of what we’ve started,” said Rowan Sones, corporate vice president of Microsoft’s Edge and Platform Group. Dr. “But they can imagine it in any kind of responsible form factor to cover any design in the world.”

Percept customers will have access to Azure’s cognitive services and machine learning models, and Percept devices will automatically connect to Azure’s IoT hub. Microsoft says it is working with silicon and equipment manufacturers to create an ecosystem of “intelligent edge devices that are certified to run with the Azure Percept platform.” Over the next few months, Microsoft plans to certify third-party devices for inclusion in the program, which would ideally allow customers to accept their proof-of-concept and easily place them on a proven device.

“Anyone who builds a prototype with one of our development kits, if they buy a certified device, they don’t have to do any extra work,” said Christa St. Pierre, product manager at Microsoft’s Azure Edge and Platform Group. St. Pierre further noted that all components of the platform must comply with Microsoft’s responsible AI policy – and undergo extensive security testing.