Machine Learning

Bringing machine learning (ML) to IoT applications reduces bandwidth requirements, saves power, and increases a device’s ability to make smarter decisions. Silicon Labs supports machine learning in all Series 1 and Series 2 wireless SoCs including newly released BG24 and MG24 products with built-in AI/ML hardware accelerator.

Silicon Labs Can Help You Bring ML to the Tiny Edge

ML at the Edge

To better navigate the need for lower power consumption and size constraints in consumer devices, small, integrated development solutions are needed. 

 

ML for Any Wireless Application

Bring ML to any application with a full portfolio of multiprotocol SoCs, a wide selection of development tools, and extensive expertise across wireless standards. 

 

ML as a Feature

Machine learning offers device manufacturers across industries and applications a way to introduce feature-rich, low-power products that stand out in the market. 

ML Development Tools for Anyone

Silicon Labs offers ML development tools for any level of experience and suited to your specific application. Depending on the tool, different sets of developer skills are required. To help determine which tool, the developer skills are grouped into three categories: ML Solutions, ML Explorer, and ML Experts



ML Solutions

Developers at this level require very little, if any, experience with ML applications and look for solutions focused on their specific use case that they can integrate into their current application. The tools offered at this level will focus on using ML as a methodology but do not require any ML experience.



ML Explorer

An ML Explorer is an experienced embedded developer familiar with ML concepts but might be working on their first ML project or are exploring how ML can help them differentiate their product. Developers at this level are interested in a tool that offers end-to-end coverage of the workflow or prefer GUI-based tools over code-based solutions.



ML Expert

An ML Expert is someone with extensive experience working on ML projects and who is familiar with TensorFlow and Python. These developers understand how to pre-process raw data and attenuate the key elements, know how to create the proper network of convolutional computations, and how to interpret the constant output of stochastic information from inferencing.



Application Examples and Getting Started Resources

Below are some of the most common use cases for ML application development, as well as access to technical documentation that can help you select the software and tools that are most applicable. You will also find demos, tutorials, and examples according to your specific application and level of ML experience.

 

Sensor signal processing is the use of low data rate sensors including accelerometers, gyroscopes, air quality sensors, temperature sensors, or pressure sensors. This makes it possible to extend machine life cycles, avoid down time, and reduce cost with preventive maintenance.

Audio pattern matching uses microphones to detect a very wide range of non-speech related sounds including squeaky bearings, breaking glass, or running water. These features make it possible to bolster in-home security with glass break detector, scream, and shot detection.

Voice commands are a specific sub-set of audio patterns that are the recognition of single words, which is also sometimes referred to as keyword spotting. Make a smart home a responsive home by turning lights on/off with AI/ML keyword detection.

Make smart devices seeing devices by wake-up upon object detection, presence detection, people counting, and more.

Getting Started With ML Software

Silicon Labs offers ML development tools for any level of experience, and we’ve curated our resources based on level of expertise:

Edge Impulse for ML Explorers

Edge Impulse is ushering in the future of embedded ML by empowering developers to create and optimize solutions with real-world data. Silicon Labs is making the process of building, deploying, and scaling embedded ML applications easier and faster than ever, unlocking massive value across every industry, with millions of developers making billions of devices smarter. 

SensiML for ML Explorers

SensiML offers AutoML embedded code generation software for implementing AI at the IoT edge. The SensiML Analytics Toolkit supports rapid data collection, ML classification, and optimized firmware code generation, and built-in automation reduces development time and cost by generating optimized edge AI sensor algorithms in a fraction of the time.

TensorFlow for ML Experts

For ML development experts, we offer native support of TensorFlow Lite for Microcontrollers natively for all Series 1 and Series 2 wireless SoCs, and offer two options to create models:

Software for ML Solutions

For developers with little or no ML experience, Silicon Labs has a partnership with Sensory for keyword and wake-word applications, and Micro.ai for anomaly detection.

AI/ML Partners

Edge Impulse is the leading development platform for embedded machine learning, free for developers, and used by over 1,000 enterprises worldwide.

SensiML pioneered software tools simplifying the development of TinyML code for IoT sensor applications.

Sensory Inc. creates a safer and superior UX through vision and voice technologies widely deployed in consumer electronics applications.

MicroAI™ is an endpoint-based artificial intelligence and machine learning engine that lives directly on a device.

Getting Started With ML Hardware

Our EFR32xG24 Development Kit allows developers to load and run example projects on a target device. This development kit runs embedded applications that use the TensorFlow engine and have an integrated ML model. All of the software described above will runs on this development kit.

EFR32xG24 Dev Kit (xG24-DK2601B)

The EFR32xG24 Dev Kit is a compact, feature-packed development platform. It provides the fastest path to develop and prototype wireless IoT products. The development platform supports up to +10 dBm output power and includes support for the 20-bit ADC as well as other key features such as the xG24's AI/ML hardware accelerator.

Close
Loading Results
Close