Edge AI Innovator Femtosense Brings Affordable, Off-the-Shelf Hearing Aids, Other Wearables Closer to Reality with First Silicon

At CES 2023, January 5-8, the company will demonstrate compelling audio applications with highly efficient, real-time artificial intelligence (AI) processing with low power, low cost requirements

Edge AI innovator Femtosense announced the arrival and successful testing of its first silicon, substantiating the start-up’s assertions that it will deliver significantly more efficient, lower cost and real-time AI processing for valuable and affordable consumer electronics. At the Consumer Electronics Show (CES) in Las Vegas January 5-8, Femtosense will demonstrate enhanced audio applications built with the new AI processor such as noise suppression and speech enhancement technologies for hearing aids, earbuds and other wearables.

Recommended AI: “Bitcoin Has No Intrinsic Value”. Then What Gives Bitcoin Value?

Edge #AI Innovator #Femtosense brings affordable, off-the-shelf hearing aids, other wearables closer to reality with first silicon #artificialintelligence

“We now have what consumer electronics manufacturers have been asking for: real silicon they can test to build compelling, highly efficient and affordable applications that just haven’t been possible until now,” said Sam Fok, Ph.D., CEO and co-founder, Femtosense. “We are already working with partners and customers to evaluate and develop products that enable unprecedented features for small-state battery-powered wearables down to home security, televisions and even audio in vehicles, all with AI that requires much less power for even greater efficiency.”

Imec says Femtosense chip is a first-pass success at the silicon manufacturing level and proven working functionality deployed at the product level.

Until Femtosense, AI has required excessive amounts of power and been too costly to deliver sophisticated AI applications outside of datacenters or flagship phones. Leveraging sparse mathematics, Femtosense’s AI processing strips away the unnecessary work in AI to significantly improve efficiency.

To build the chip, Femtosense teamed with Imec.IC-link, a complete ASIC (application-specific integrated circuit) solutions provider that can manage the full product lifecycle.

Recommended AI: Top 10 Countries and Cities by Number of CCTV Cameras

“Femtosense and IC-link engaged in a very early stage of the ASIC development,” said Yiyi Wang, head of Imec.IC-link North America. “We believed in Femtosense’s technology that could be implemented for many edge AI applications that require the combination of latency, low power and low cost. IC-link provided foundry tape-out, silicon manufacturing and backend packaging services. We are impressed with Femtosense’s design and engineering capabilities and how fast this team can deploy the design with all the features and turn the first silicon around within months. We are very proud to hear the great feedback from their consumer electronics manufacturers.”

In October, Femtosense raised $8 million Series A funding led by Fine Structures Ventures. The company is using the funds to accelerate the development and delivery of the chips. Femtosense expects volume production in 2023.

Recommended AI: How is Artificial Intelligence (AI) Changing the Future of Architecture?

[To share your insights with us, please write to sghosh@martechseries.com]