Silicon Labs and Edge Impulse Accelerate Machine Learning Applications

Beijing, China – February 3, 2021 – Silicon Labs (also known as Silicon Labs, NASDAQ: SLAB), a leading supplier of chips, software and solutions dedicated to building a smarter, more connected world, announced a partnership with leading edge Edge Impulse, an on-device machine learning (ML) development platform, has teamed up to enable rapid development and deployment of machine learning applications on Silicon Labs EFR32 wireless system-on-chip (SoC) and EFM32 microcontrollers (MCUs). The Edge Impulse tool enables sophisticated motion detection, sound recognition, and image classification on low-power, memory-constrained remote edge devices.

Research shows that 87% of data science projects never reach production, often due to artificial intelligence (AI)/machine learning challenges. With this new collaboration between Silicon Labs and Edge Impulse, device developers can generate and export machine learning models directly to the device or Simplicity Studio (Silicon Labs’ integrated development environment) at the click of a button, in digital Machine learning can be implemented in minutes.

“Silicon Labs believes that our efforts to infuse machine learning into edge devices will make the Internet of Things smarter,” said Matt Saunders, vice president of IoT at Silicon Labs. “Edge Impulse provides secure, private, and easy-to-use tools that enable machine learning in It saves developers time and money and brings amazing new user experiences to real-world business applications ranging from predictive maintenance and asset tracking to monitoring and people detection.”

Through integrated deployment in Simplicity Studio, Edge Impulse enables developers to quickly create neural networks on a variety of Silicon Labs products for free. By embedding state-of-the-art TinyML models in EFR32 and EFM32 devices such as MG12, MG21, and GG11, the solution enables the following:

· Real sensor data collection and storage

Advanced signal processing and data feature extraction

· Machine Learning

· Deep Neural Network (DNN) model training

· Optimize embedded code deployment

Edge Impulse tools can also leverage Edge Impulse’s Edge Optimized Neural (EON™) technology to optimize memory usage and inference time.

Zach Shelby, co-founder and CEO of Edge Impulse, said: “The applications of embedded machine learning in industrial, enterprise and consumer are endless. Integrate machine learning with Silicon Labs’ advanced development tools and multi-protocol solutions. , will bring customers excellent wireless development opportunities.”

Support for Edge Impulse is ready for Silicon Labs’ Thunderboard Sense 2 and wireless SoCs and MCUs. For more information, visit: silabs.com/solutions/artificial-intelligence-machine-learning. Edge Impulse will be hosting a hands-on workshop at the tinyML Summit (March 22-26, 2021) to learn more about the AI/ML capabilities of the Silicon Labs platform. The first 250 workshop registrants will receive a free Silicon Labs development kit to use during the event.

About Edge Impulse

Edge Impulse is the leading embedded machine learning development platform used by more than 1,000 enterprises worldwide in more than 10,000 machine learning projects. Our mission is to enable the ultimate development experience for machine learning at scale on embedded devices for sensing, audio and computer vision. From onboarding to production machine learning operations (MLOps) in 5 minutes, we provide highly optimized machine learning capabilities that can be deployed on a variety of hardware from MCUs to CPUs.

About Silicon Labs

Silicon Labs (also known as Silicon Labs, NASDAQ: SLAB) is a leading provider of silicon, software and solutions dedicated to building a smarter, more connected world. Our award-winning technologies are shaping the future of IoT, Internet infrastructure, industrial automation, consumer electronics and automotive markets. Our world-class engineering teams create products focused on performance, energy efficiency, connectivity and simplicity.

The Links:   EVM31-060 1SI50A-050 SKM200GAL126D