NanoEdge AI Studio v5 is the first AutoML tool for STM32 microcontrollers capable of generating anomaly data out of typical logs, thanks to a new feature we call Synthetic Data Generation. Additionally, the latest version makes it easier to create datasets, configure AI libraries, visualize information, run data logging sessions, and more. In a nutshell, as data remains at the center of the AI revolution, NanoEdge AI Studio v5 not only helps process but generate quality data. If machine learning algorithms are only as good as the data that shapes the neural network, then NanoEdge AI Studio v5 has just made creating AI at the edge significantly easier and more accurate by solving one of the biggest development challenges.
Table of Contents
Would the real AI challenges please stand up?
The unfair burden
The question is no longer what category of embedded systems uses AI, but rather which ones don’t. Avid readers of the ST Blog would be hard-pressed to name one market that isn’t even minimally influenced by smart sensors, predictive maintenance, data analysis, automated decision-making, and more. The problem is that engineers must now be experts in so many new fields. In a typical application, teams must understand data science, neural network architectures, inference optimization, and other related concepts. Before they can even create an application, they must gather data, select the appropriate models, work with new tools, and even programming languages unrelated to their primary activities. In other words, engineers must venture out of their comfort zone.
The exorbitant costs
The crisis is real. According to a [2020 report by Anaconda], dedicated data scientists spend more than 65% of their time loading, cleaning, and visualizing data. Even if machine learning algorithms at the edge are less complex, they are still costly to create from scratch. Smaller companies may, therefore, feel pressure to adopt AI at the edge, but lack the financial resources to hire talents and invest in necessary resources, which can rapidly overwhelm existing teams. For instance, a speech recognition project shared by the Barcelona Supercomputing Center required logging 2,250 hours of speech “from approximately 20 thousand distinct speakers.” That’s a scale that’s out of reach for nearly all small to medium-sized enterprises.
The data impasse
Another issue is the gathering of quality data. Industrial applications sometimes find it impossible to replicate certain conditions. For instance, no company knows how a motor will break, but they must find a way to replicate that situation, or no one will have data to train AI models. It can lead engineers down an impossible path where they have to find a way to anticipate how to detect a failure without knowing what it will look like. And trying to imagine an adverse event, or simulating one, can be time and resource-consuming with very little to show for if the real-world event is significantly different from the training data.
Stayin’ Alive: 3 practical solutions from NanoEdge AI Studio v5 to fight the office night fever
Synthetic Data Generation
It’s precisely those issues that drove ST to work on an application like NanoEdge AI Studio v5. For instance, Synthetic Data Generation solves the challenge of replicating anomalies by analyzing nominal datasets and then simulating an issue by adding noise or a vibration drift. After working on anomaly detection for more than five years, since the first release of NanoEdge AI Studio, ST has acquired expertise that enables us to anticipate abnormal behaviors and create synthetic data based on real-world use cases we’ve encountered. It explains why we are the only ones to offer this feature for time series applications. The tool currently focuses on vibration data, but future releases will target other applications.
Feature Importance
To further assist developers struggling with optimization, ST provides Feature Importance. The new functionality utilizes statistical analysis to identify the sensors or data columns that have the most significant impact on the AI algorithm, thereby prioritizing them to minimize the footprint of the machine learning application, with minimal to no effect on inference. Among other things, it provides a direct solution to the collection and storage problem, as developers can spend less time cleaning and selecting data and fewer resources to store it. It’s essentially a data scientist for all, thereby alleviating the burden that developers face.
Widget-based visualization
NanoEdge AI Studio v5 also offers new visualization tools to simplify data processing, even for users with limited experience in data science. Concretely, the new widget allows users to customize various layouts to optimize their workflow. For example, it’s possible to display a feature chart, raw signals, and model performance views side by side, which quickly helps explain how the data affects the machine learning algorithms without developers having to write a single line of code. The ST software even offers the ability to choose the number of columns, the number of widgets per column, resize them, and determine which widgets appear where. Put simply, it’s the most customizable version of NanoEdge AI Studio ever shipped.
“Beat It (Beat It), Beat It (Beat It)”, No one is defeated with ST Edge AI Suite
NanoEdge AI Studio as a reference
NanoEdge AI Studio is often the first step in numerous AI journeys, as engineers bring their data and turn it into something meaningful that can serve a machine learning algorithm with just a few clicks. In fact, a recent study by the 1 compared the most popular AutoML tools and found that NanoEdge AI Studio obtained the “best results”. According to their findings, the ST software was the only one even to surpass the performance of a custom model. Hence, while the researchers acknowledge the ease of use and attractiveness of tools like NanoEdge AI Studio, the ST utility stands out thanks to its capabilities. It was also the only genuinely free tool on the list.
ST Edge AI as an ecosystem
NanoEdge AI Studio also has another critical benefit: it is part of the ST Edge AI Suite, a repository of free software tools, use cases, and documentation. The hub helps developers create AI for the Intelligent Edge, regardless of their experience level. For instance, STM32Cube.AI converts neural networks from popular frameworks, like TensorFlow, and optimizes them for STM32 MCUs. The ST Edge AI Developer Cloud serves to benchmark and validate models against real ST hardware thanks to its Board Farm. ST even has an STM32 Model Zoo so users can jumpstart their project by choosing a pre-trained model. And our high-speed data logging toolkit captures data at up to 6 Mbit/s from sensors to streamline data acquisition.
Consequently, while many engineers begin in NanoEdge AI Studio, many also utilize the ST Edge AI Suite and its included tools to expand the scope of their work. For example, those who wish to run AI algorithms on sensors will gravitate toward ST AIoT Craft, the first cloud solution for ST’s machine learning core and MEMS Studio. ST’s extensive ecosystem thus highlights the need for developers to have specialized tools that can provide them with the expertise they may not have or the ability to hasten development to meet the new demands brought on by machine learning at the edge.
- Montes-Sánchez, J. M., Fernández-Cuevas, P., Luna-Perejón, F., Vicente-Diaz, S., & Jiménez-Fernández, Á. (2025). No-Code Edge Artificial Intelligence Frameworks Comparison Using a Multi-Sensor Predictive Maintenance Dataset. Big Data and Cognitive Computing, 9(6), 145. https://doi.org/10.3390/bdcc9060145 ↩︎
