Azure Percept: Edge intelligence from silicon to service

In today’s fast-paced world, there is often a need for instantaneous, real-time responses. Companies are finding the computing needed to support this 24/7 mentality generates staggering amounts of data, often from devices out in the physical world. This influences a shift in priorities as companies look for technology that gives them the flexibility to innovate anywhere in their environment. The biggest trend: the evolution of cloud strategies to include edge and hybrid investments.

This hybrid cloud journey gives companies the flexibility to easily handle future growth and technology needs. Add in the movement toward the intelligent edge, and companies can extend compute and artificial intelligence (AI) to the edge of a network to unlock new business scenarios. However, edge AI technologies often need to be architected from the ground up for every solution and require an almost data-scientist-level of skill and experience to develop, deploy, and operate.

Which is why we are introducing you to Microsoft Azure Percept, a comprehensive, easy-to-use platform with added security for creating responsible edge AI solutions. With Azure Percept, we have simplified AI on the edge from silicon to service. As we share during Microsoft Ignite, you can start prototyping in minutes, getting real-time edge AI insights for quick decision-making and analytics when and where the action occurs to help speed the development of your solution.

Comprehensive edge AI platform

Azure Percept focuses on extending AI to the edge with a single, end-to-end platform. We integrated hardware accelerators with Azure AI and Azure IoT services and designed them to be simple to use and ready to go with minimal setup. We also designed Azure Percept in accordance with Microsoft responsible AI principles to help ensure all of these services are streamlined to create edge AI solutions that can adapt to customers’ needs—while helping ensure customers’ trust as the edge evolves.

Azure Percept DK by ASUS is a pilot-ready development kit with a carrier board, mounting tools, and Azure Percept Vision, a camera-enabled system on module (SoM) for rapid scenario-based customization. Azure Percept Vision adds camera and vision functions, optimized for industrial and retail environments, to power shelf-stock analytics, people and object detection, and loss prevention among other uses. This development kit is deeply integrated to work with Azure AI, Azure Machine Learning, and Azure IoT management services right out-of-the-box.

Your development kit can be extended with Azure Percept Audio, an audio hardware accelerator with a preconfigured four-mic linear array, to create customized speech functions that react to spoken commands, keyword spotting, and far field speech to local listening devices.

Simple to build and manage

For those looking for solution management and maintenance for devices, workloads, and high-value AI and machine learning models—whether citizen or advanced developers—Azure Percept offers an end-to-end blueprint for building and training edge AI solutions. Azure Percept Studio hosts a collection of fully managed Azure services with added security and integrated for simple-to-use device provisioning and edge AI lifecycle management, giving developers a full suite of services designed to be deployed on edge hardware.

Using Azure Percept Studio, developers of all skill levels can explore the library of pre-built Azure AI models or build custom no-code models of their own. Azure Percept works out-of-the-box with Azure Cognitive Services, Azure Machine Learning, Azure Live Video Analytics, and other Azure services to deliver vision and audio insights in real-time.

Designed with modern security in mind

Azure Percept security measures are built into an edge AI solution using a Zero Trust security model so you can help safeguard your sensitive AI models, data in transit, and at rest. It starts with security measures built-in from silicon to service, helping to ensure security posture is monitored and governed, and data protection is part of every edge AI solution. Azure Percept hardware accelerators incorporate a hardware root of trust, enabling strong device identity.

Azure Percept also integrates with Azure IoT Hub Device Provisioning Service (DPS) to provision devices to Azure with hardware-backed identity, and it uses Wi-Fi Zero Touch Provisioning with Wi-Fi Easy Connect™.

Source: microsoft.com