ST Edge AI Suite Supports Development of Solutions Utilizing ST Hardware and Embedded AI Tools
Plans to integrate existing AI ecosystems such as AWS, software tools, evaluation, and development kits

STMicroelectronics (ST) is offering the ST Edge AI Suite to help customers easily develop AI solutions by taking full advantage of ST’s hardware (general-purpose and automotive microcontrollers, microprocessors, and smart sensors) and related tools optimized for embedded AI.

ST announced on the 7th that it provides a comprehensive ecosystem that supports corporate product innovation with Edge AI.

With the announcement of ST Edge AI Suite, a set of integrated software tools that can be used free of charge with ST hardware, customers can now go one step further and start designing and deploying billions of connected, autonomous things with artificial intelligence locally.

A key component of the ST Edge AI family is the ST Edge AI Core, which integrates all the software and tools engineers need at each stage of a project.

ST Edge AI Core imports ML and NN algorithms from the most popular ML frameworks, provides detailed analytics, optimizes the algorithms for the selected device (sensor, MCU, MPU), and compares them to the original model.After verification, the final derived embedded AI solution is mapped to the relevant device.

Benchmark the same algorithms on different platforms, either purely SW or leveraging specific HW accelerators, and evaluate accuracy and inference time with just a few clicks.

We will leverage the existing AI ecosystem of machine learning frameworks and key partners (NVIDIA, AWS, etc.), while expanding and integrating various software tools, evaluation and development kits released over the past decade.

“We are moving towards a world where billions of autonomously connected things will provide users with value and convenience in every aspect of consumer life and enterprise productivity,” said Jean-Marc Chery, President and CEO, STMicroelectronics. “To achieve this, AI algorithms must run at the Edge, both in the cloud and on-device, across devices that include smartphones, connected personal devices, smart home and building control systems, industrial equipment, automobiles, and more.”

“ST products are already key to all of these devices, but when combined with the industry-leading suite of free software that we are announcing today, they are truly differentiated,” he continued. “This combination will enable the transition to a more intelligent Edge, enabling customers of all sizes to more easily deploy Edge AI and build their vision of a connected future, leveraging ST’s hardware portfolio.”

A preview of the ST Edge AI product family features was unveiled at ST's online Edge AI Summit held on the 6th.

ST's solutions are designed for embedded developers who want to build optimized machine learning models and data scientists who want to run models on embedded devices. It will be of great help to scientists, product designers and manufacturers who are trying to redefine product excellence.

“Free access enables customers large and small to pool resources and knowledge in a community-driven manner,” ST said. “This product family will accelerate the transition of both tools and users, centered around the broader Edge AI community.”

Meanwhile, the first release of the ST Edge AI family is expected in the first half of 2024.