Science & Technology

Edge Impulse Earns $ 34 Million As TinyML Market Continues to Grow

Join the game leader with GamesBeat and Facebook Gaming to participate in the 2nd GamesBeat & Facebook Gaming Summit. GamesBeat: To Metaverse 2 from January 25th to 27th, 2022. See event details.


As companies increasingly pilot AI technology, small machine learning (TinyML) is emerging as a great way to reduce the resources required for deployment. TinyML is a machine learning technique that can be implemented in low energy systems such as sensors to perform automated tasks. This technology is still a very clear AI, but it uses low power and costs and often does not require an internet connection.

TinyML’s applications run the full range, but the most popular ranges are factories, retail and agriculture. In the manufacturing industry, TinyML can prevent downtime by warning workers to perform preventive maintenance based on the condition of their equipment. Also in agriculture, TinyML monitors livestock vital signs to help identify early signs of illness.

Many start-ups offer products designed to help companies implement TinyML solutions, but most notably. Edge impulse.. Launched in 2019, Edge Impulse provides a platform and services for developing devices that leverage embedded AI and machine learning. Approximately 30,000 developers from thousands of companies, including Oura, Polycom, and NASA, use Edge Impulse solutions to build industrial, logistics, consumer, and health solutions and create over 50,000 custom projects. Claims to have done.

Advances in TinyML

Edge Impulse was founded two years ago by Jan Jongboom and Zach Shelby. Jongboom previously contributed code to Mozilla’s now obsolete operating system, Firefox OS, leading some developer evangelism on Arm’s Internet of Things (IoT) platform. Shelby comes from an investment background and has been a member of Petasense’s board of directors and proptech startup Cubi Casa.

Edge Impulse allows developers to collect or upload training data from their devices, label the data, train the model, and deploy and monitor the model in a production environment. The platform supports the development of machine learning for sensors, audio and computer vision, specializing in TinyML industrial applications such as predictive maintenance, asset tracking, monitoring and sensing.

“The accuracy normally used to evaluate the performance of a machine learning model is only a small part of the story. You need to know the strengths and weaknesses of the model and when to miss an event or cause a false positive. “There is,” Shelby told VentureBeat in an email. “”[That said,] Machine learning has great value potential for any company dealing with sensor-related data, from cost savings and improved customer service to the realization of a whole new generation of functional value. “

Above: Edge Impulse development dashboard.

Image credit: Edge Impulse

To increase the efficiency of platform-trained models, EdgeImpulse uses a compiler that compiles the model into C ++. The company claims that this can reduce RAM usage by 25% to 55% and storage usage by up to 35% compared to rival approaches.

“”[We’ve seen] Human keyword detection on wearable battery-powered devices, predictive maintenance on smart grids, gesture recognition using radar on devices, monitoring of critical refrigeration equipment in the field, field detection of eye diseases, using audio Enterprise applications such as weld quality monitoring [and] Construction and manufacturing safety monitoring using computer vision and sensors, “Shelby said. “In the early days of the pandemic, there were no customers on the sites that needed to collect data, which slowed down the pace of customers, but the business recovered very strongly.”

Year of growth

according to By 2027, for Gartner, machine learning in the form of deep learning will be included in more than 65% of edge use cases, up from less than 10% in 2021. Meanwhile, ABI Research Predict The TinyML market is expected to grow from 15.2 million device shipments in 2020 to 2.5 billion units in 2030.

Reflecting the growth of a wide range of segments, Edge Impulse reports that the platform’s developer base has quadrupled last year and annual recurring revenue has tripled. In related news, the company today raised $ 34 million in a Coatue-led Series B round with the participation of Canaan Partners, Acrew Capital, Fika Ventures, Momenta Ventures and Knollwood Investment Advisory, doubling its valuation. Announced that it has reached $ 234 million. Total capital has been raised to over $ 54 million.

According to Shelby, the new funding will be used to expand a team of about 40 people at Edge Impulse and a network of hardware partners already including Nvidia, Texas Instruments, Syntiant and Synaptics. “Use this money to accelerate even faster, significantly expand the developer ecosystem to 100,000 developers by the end of 2022, grow the solution engineering team rapidly, and make customers more successful and hard. We will expand the hardware ecosystem and allow us to invest in new R & D creation. We can make machine learning of sensors, audio and computer vision more efficient, “he added.

Edge Impulse conflicts with startups such as CoCoPie, Neural magic, NeuReality, Deci, CoCoPie, When DeepCube, Above all. FogHorn Is one of the closest direct competitors, offering a variety of edge intelligence software for industrial and commercial applications. Existing companies such as Microsoft, Amazon, and Google also provide services for edge AI development via their respective cloud platforms (such as IoT Greengrass of Amazon Web Services).

VentureBeat

VentureBeat’s mission is to become a Digital Town Square for technical decision makers to gain knowledge about innovative technology and trading. Our site provides important information about data technologies and strategies to guide you when you lead an organization. We encourage you to become a member of the community and access:

  • The latest information on the subject you are interested in
  • Newsletter
  • Gated sort reader content and discounted access to valuable events such as: 2021 transformation: learn more
  • Network function etc.

Become a member

Edge Impulse Earns $ 34 Million As TinyML Market Continues to Grow

https://venturebeat.com/2021/12/09/2742993/ Edge Impulse Earns $ 34 Million As TinyML Market Continues to Grow

Back to top button