Syntiant Corp. is one of the most closely watched local tech upstarts, and interest in the Irvine chipmaker is likely to grow in the future.
Founded in 2017, it’s already raised more than $120 million to date, ramped up production and delivered more than 20 million chip units.
“We’re doing quite nicely,” says Kurt Busch, co-founder and CEO.
The company’s chips detect voices for products such as Amazon Alexa including custom wake words and commands, and are now being extended for more audio, visual and sensing detection.
Busch says Syntiant is already in the
Ring Alarm Glass Break Sensor system. The company has also developed uses for smoke detection and fire detection.
Syntiant is backed by key investors including Intel Capital and Microsoft’s M12.
Other Syntiant investors include Applied Ventures, Robert Bosch Venture Capital, the Amazon Alexa Fund and Atlantic Bridge Capital.
Busch sees continued growth at a “rapid rate” for always-on voice, sensor and vision applications. Its chips use only tiny amounts of power in a whole array of devices and are ultra-quick to react.
The chips can even be used to aid in license plate recognition and the detection of door tampering.
Public Plans
Syntiant’s steady progress may bring up again the idea of going public. Fellow chip startup Mobix Labs in Irvine last month filed regulatory documents to go public via reverse merger; for more, see the April 24 print edition of the Business Journal.
While Busch sits on the Mobix Labs board and helped to bring about the public listing thanks to a prior business connection, he says that the time isn’t right for Syntiant just yet.
“The company (Syntiant) has always been designed to be a public company,” Busch told the Business Journal on April 13. “We’ll be seriously considering that a few years out.”
Mobix Labs
There are differences in the two upstart semiconductor companies’ strategies.
For one, Mobix Labs says its strategy “relies in part on acquisitions of companies, assets or technologies to create growth and increase cash flow.”
An M&A strategy is generally considered to be more appropriate for a public company than it is for a private company.
“Syntiant’s strategy is really organic growth,” Busch says. “That we can do quite well as a private company.”
Syntiant is enjoying a hefty cash position, six years into operations.
“We’ve raised a little over $120 million to date and that gives us a quite long runway. We’re currently ramping revenue. We have over two dozen customers.”
He declines to share revenue figures, saying: “That’s the advantage of being private.”
Size is another point.
Syntiant currently has a little over 80 employees, with about 20 people in Silicon Valley.
“The majority of Syntiant’s team is at headquarters in Irvine, and we also have a team in Silicon Valley,” according to Busch.
He said the company needs to have a “level of size that is appropriate” for a public listing.
“Today we’re still too small,” but he’s confident the company will grow to the size needed for a public listing.
The Syntiant chips use artificial intelligence at the “edge” rather than being routed through cloud computers.
“The expectation is that AI at the edge is a huge but new market.”
Syntiant said two years ago it supports “OK Google” and “Hey Google” hotwords with its chip units.
Product Accolades
Flying the flag is a key part of the company’s plans.
“We actually have done three trade shows since the beginning of the year,” Busch says.
Syntiant’s NDP200 Neural Decision Processor was named “Best Product of the Year” by the tinyML Foundation.
The company was also named 2022’s “Start-Up to Watch” during the Global Semiconductor Alliance Awards late last year.
The company showed off its products at the giant CES electronics and tech show in Las Vegas; the ISC West security conference; and at the Embedded World 2023 conference, held from March 14-16 in Nuremberg, Germany.
“In all three of those we showed both our audio event detection solutions as well as our computer vision solutions,” he says. “All of those things with Syntiant are tied directly to the AI at the sensor, so it can be done without having to go to the cloud.”
He says AI use at the devices also increases privacy and speed and lowers cost.
Computer Vision
Busch said in December the company has developed uses for smart door locks, battery management for electric vehicles, in addition to sound detection such as glass break-ins.
“We are currently being used in products as small as hearing aids to as large as automobiles,” the CEO says.
Syntiant developed deep learning models for video conferencing deployed on a partner’s processor.
The company’s expansion now positions it well for other applications such as home security with smart cameras and doorbells, according to the CEO.
“We’ve rolled out a computer vision product line,” according to Busch.
Computer vision is the subcategory of artificial intelligence that focuses on building and using digital systems to process, analyze and interpret visual data.
Syntiant says that with its computer vision models for smart doorbells, for example, “you can actually discern between a car, a pet and a person.”
“We see that computer vision perhaps doubles our addressable market, maybe even more,” according to Busch.
