Lattice sensAI Solution Portfolio Delivers Next-Generation AI Experiences at the Edge

The AI ​​and machine learning revolution continues to sweep across multiple application areas, especially network edge applications. Network edge devices such as security cameras, robots, industrial equipment, client PCs and even toys can now support AI/ML capabilities, bringing new features and experiences to users. Industry analyst firm ABI Research said that the network edge AI chipset market "has experienced strong growth in the past and is expected to continue to grow to $71 billion by 2024, with a compound annual growth rate of 31% from 2019 to 2024. Such strong growth is due to the migration of AI inference workloads from the cloud to the network edge, especially in smartphones, smart homes, automobiles, wearable devices and robotics industries."

However, making client computing devices "intelligent" brings new challenges to product design. As AI/ML is an emerging technology, many OEMs do not have enough team experience or time to design solutions from scratch. On the other hand, the algorithms used to train client computing devices are developing rapidly, so developers are also looking for field-upgradeable AI/ML solutions. But for many developers of AI/ML applications at the edge of the network, the most critical issue is how to provide enough processing performance to implement AI/ML functions in battery-powered devices?

To address these issues, application developers and OEMs need flexible hardware and software solutions to enable AI/ML experiences at low power. Since 2018, the Lattice sensAI™ solution stack has helped Lattice customers add AI/ML capabilities to new and existing product designs.

We are pleased to announce that the latest version (v4.1) of our award-winning sensAI solution collection now includes a user experience reference design roadmap that brings AI/ML capabilities to client computing devices such as laptops. The COVID-19 pandemic has led to a significant increase in the number of people using video conferencing applications to work remotely and stay connected with friends and family. The reference designs included in the latest sensAI version leverage the vision and sound sensors in client devices to provide high-value user experiences such as instant-on , person detection , attention tracking , privacy protection , and video conferencing , while maintaining low power operation to maximize battery life.

Lattice sensAI Solution Set v4.1 helps developers use sensors and AI/ML inference to deliver new and optimized user experiences for client computing devices

By adding AI/ML support to their device designs using Lattice sensAI, OEMs can reap a number of benefits:

Lattice will add more new experiences to its client computing AI roadmap in future releases of sensAI.

sensAI also supports the latest member of the Lattice Nexus™ family, the Lattice CertusPro™-NX device, which improves performance and accuracy for applications such as high-precision object detection and defect detection in automated industrial systems. The solution set also supports a new hardware platform, including an onboard image sensor, two I2S microphones, and expansion connectors for adding more sensors, to facilitate the development of voice- and vision-based machine learning applications.

On the software side, sensAI has updated its neural network compiler to support Lattice sensAI Studio , a GUI-based tool with a library of AI models that can be configured and trained for a variety of mainstream application scenarios. sensAI Studio now supports AutoML, which can create machine learning modules based on application and data set targets. Some models based on the Mobilenet machine learning inference training platform are optimized for the latest Nexus series product, Lattice CertusPro™-NX. sensAI is also compatible with other widely used machine learning platforms, including the latest versions of Caffe, Keras, TensorFlow, and TensorFlow Lite.

The latest version of sensAI includes hardware and software updates that make it easier to add AI/ML support to more network edge devices (updates are shown in the yellow box)

Our entire solution portfolio will be updated regularly, so please Sign up for a Lattice account to receive news about our latest products and solutions.