STMicroelectronics Takes Up AI To New Level With Its Signatory STM32 Neural-Network Developer Toolbox

STMicroelectronics recently took leverage from its STM32 family of microcontrollers for expanding its STM32CubeMX ecosystem with new AI (Artificial Intelligence) features. It will take care of product developers all over the world allowing STMicrocontrollers to maintain its crowning position in terms of microcontrollers.

Fig. 1: Representational Image of STM32CubeMX.AI

Artificial Intelligence makes use of artificially created neural networks capable of classifying data signals from image sensors, microphones, environmental sensors, vibration sensors, as well as motion sensors.

As the President of Microcontrollers and Digital ICs Group at STMicroelectronics, Claude Dardanne puts it, “ST’s new neural-network developer toolbox is bringing AI to microcontroller-powered intelligent devices at the edge, on the nodes, and to deeply embedded devices across IoT, smart building, industrial, and medical applications.”The STM32Cube.AI will help developers transform pre-trained neural networks into C-code that calls the function in the optimized libraries that work on STM32 MCUs.

Features

The STM32Cube.AI comes embedded with some readymade software function packs that include sample code for audio scene classification and human activity recognition. Such sample codes can be used immediately with the STSensorTile reference board as well as the ST BLE Sensor mobile app.

It also comes with some add-on features like engineering services and dedicated Machine learning and AI STM32 online community. ST will showcase the apps developed with the help of STM32Cube.AI in a private suite at the CES 2019 in Las Vegas. This pack can be downloaded in the ST’s STM32CubeMX MCU configuration as well as software based code-generated ecosystem.