STMicroelectronics, Global semiconductor company, has launched a new series of microcontrollers, for the first time integrated with embedded artificial intelligence (AI) and accelerated machine learning (ML) capabilities to boost AI at the edge. The company believes the chip series enables cost-sensitive, power-conscious consumer and industrial products to provide high-performance features leveraging computer vision, audio processing, sound analysis and other algorithms that were previously beyond the capabilities of small embedded systems.
Also Read: Qualcomm Launches Wireless Networking Platform With Wi-Fi 7 and Edge AI
STM32N6 MCU with AI and ML Capabilities
STMicroelectronics says the STM32N6 microcontroller (MCU) series is the company's most powerful to date, and the first to embed ST's proprietary neural processing unit (NPU), the Neural-ART Accelerator, delivering 600 times more machine-learning performance than a high-end STM32 MCU today.
Early Customer Adoption
The STM32N6 has been available to selected customers since October 2023 and is now available on a commercial scale. LG Electronics, Lenovo Research, Alps Alpine, Carlo Gavazzi, Meta-bounds, and Autotrak are named as early customers and have shared their experiences with the new MCU series.
"We are on the verge of a significant transformation at the tiny edge. This transformation involves the increasing augmentation or replacement of our customers' workloads by AI models. Currently, these models are used for tasks such as segmentation, classification, and recognition. In the future, they will be applied to new applications yet to be developed," said Remi El-Ouazzane, President of Microcontrollers, Digital ICs and RF Products Group (MDRF) at STMicroelectronics.
Neural-ART Accelerator
"The STM32N6 is the first STM32 product to feature our Neural-ART Accelerator NPU. It will utilise a new release of our unique AI software ecosystem package. This marks the beginning of a long journey of AI hardware-accelerated STM32, which will enable innovations in applications and products in ways not possible with any other embedded processing solution," Remi added.
The new machine-learning capabilities make it possible to run computer vision, audio processing, sound analysis, and more consumer and industrial applications at the edge, STMicroelectronics said in a statement on Tuesday.
Also Read: Qualcomm Partners with Indian ODMs to Boost Edge AI Deployment: Report
Edge Impulse to Enhance AI Development
To support the newly launched MCU series, Edge Impulse, a platform for building, deploying, and scaling edge machine learning models, and STMicroelectronics also announced support for the STM32N6 MCU.
In addition to its comprehensive suite of tools for building edge-capable AI algorithms (from data collection to model deployment), Edge Impulse offers multiple generative AI-based features, such as AI labelling, to accelerate the creation of edge AI solutions, as well as integrations with Nvidia's TAO and Omniverse. This complements STMicroelectronics' existing collaboration with Nvidia, the company said.
Possibilities of Edge AI Applications
According to the statement, with official Edge Impulse support, developers and enterprises can seamlessly leverage the STM32N6 and speed up the creation of a multitude of applications, including computer vision solutions for industrial and consumer applications, energy management, and anomaly detection.
"The STM32N6 is the perfect complement to the Edge Impulse platform," says Zach Shelby, CEO and co-founder of Edge Impulse. "With its powerful AI capabilities and energy-efficient design, it aligns perfectly with our mission to make edge AI accessible and impactful across every sector."
"By combining our state-of-the-art Neural-ART Accelerator with the intuitive tools from Edge Impulse, we are empowering developers to bring sophisticated AI capabilities directly to the edge," says Patrick Aidoune, General Purpose MCU Division General Manager at STMicroelectronics.
This announcement was made during the ongoing STM32 Summit, held from December 10-13, 2024.
