Synaptics integration could enable voice biometrics in low-power edge devices

Synaptics and Edge Impulse have recently entered a new partnership to facilitate the development of embedded machine learning (ML) models in power-sensitive intelligent Internet of Things (IoT)  applications.

The collaboration will see the integration of Synaptics’ Katana Ultra-Low Power Edge AI Platform within the Edge Impulse software development platform, designed to enable developers to create, train and deploy custom models for a wide range of artificial intelligence (AI) applications.

“Edge Impulse has a wealth of experience through its highly efficient platform for rapid development of ML models specifically for low power IoT applications,” explained Venkat Kodavati, Chief Product Officer at Synaptics.

“This partnership provides our customers a fast and efficient way to create and optimize custom models for the Katana platform that will significantly improve overall time to market for battery-powered connected devices at the edge.”

From a technical standpoint, the Katana system on a chip (SoC) platform offers a multi-core architecture optimized for ultra-low-power and low latency voice, as well as audio and vision applications.

With the integrated advanced Neural Network Engine, the solution supports the simultaneous inference of both computer vision and sound detection events with the ability to process both AI models in parallel.

In order to provide complete modules and further simplify system design and implementation, the Katana SoC platform also integrates Synaptics’ Wi-Fi and Bluetooth solutions.

With the new partnership and the integration of Katana’s capabilities with the Edge Impulse platform, developers will now be able to greatly speed up the deployment of low-power AI-enriched edge devices for offices, retail, factories, and smart homes. 

Possible applications for the new technology include object and pattern recognition, people detection and counting, environmental sensing, and more.

The first Katana Evaluation Kit is scheduled for release on the Edge Impulse platform by the end of the year.

The news comes months after Edge Impulse raised $15 million in a Series A investment round aimed at democratizing machine learning development for edge devices.

More recently, Synaptics acquired DSP Group in a $450 million deal.

Originally published at Edge Industry Review.

Article Topics

AI  |  biometrics  |  biometrics at the edge  |  computer vision  |  edge AI  |  machine learning  |  object recognition  |  Synaptics  |  voice biometrics