Jabra is broadening virtual horizons with artificial intelligence

Aurangzeb Khan explains how the firm is transforming videoconferencing hardware with AI and intelligent visual displays, and how this will benefit meeting participants

Elly Yates-Roberts
By Elly Yates-Roberts on 23 August 2021
Jabra is broadening virtual horizons with artificial intelligence
Jabra

Communication is more than just sound. It is made up of intonation, body language, gestures, eye contact and more. Jabra understands this and has built its new videoconferencing device to capture all forms of communication.

“When we look at the world with our eyes, we see a lot and we use that visual context to naturally understand what’s going on around us,” says Aurangzeb Khan, senior vice president of intelligent vision systems at Jabra. “While we talk, we look at the body language and how the person is responding. All of that happens so instinctively that we don’t even notice it. 

“If you look at traditional video conferencing in a group context, it didn’t work,” says Khan. “The field of view of a conventional single camera is about 70 degrees, which leads to a situation where you have long narrow conference rooms, or people are just off camera. If you’re off camera, there is this cognitive dissonance which causes people to feel that they aren’t a full and equal participant in the conversation.”  

This is why Jabra created Panacast 50. The device features three cameras that provide a 180-degree field of view. It also uses stitching technology to create a seamless visual experience that prevents the distortion of speakers and “makes it easy for your brain to focus on the conversation,” says Khan.  

Solutions like Panacast 50 will be an integral part of the workplace going forward, as many organisations are showing a preference for hybrid working, where employees may work in the office and at home.

“We believe that going forward at least one, if not many or most, meeting participants will not physically be in the room,” says Khan. “So how do we give everyone the best experience?” 

Khan believes there are three major factors to consider: voice quality, visual quality and artificial intelligence (AI).  

“Other organisations are trying to patch up their solutions to deliver all of these things, but we developed a next-generation architecture that features nine processors, three cameras, eight microphones, four speakers and two AI processors,” says Khan. “Customers want to buy an experience and the more easily that the technology fades into the background, the better.”

The Panacast 50 can also facilitate working methods that aren’t typically associated with digital solutions, such as whiteboards.  

“I like to think with my hands,” says Khan. “I’ll get up and start sketching on the whiteboard during a meeting. But for those who don’t use our technology, that can be a bad experience because they can’t see what I’m doing. With our multi-camera array, you can identify the whiteboard in it and the back-end technology will extract, rectify and present that whiteboard as if you were standing directly in front of it. 

“Our mission was to build a next-generation architecture which anticipates these needs and fulfil them in a simple and elegant way.” 

AI is a big part of how Khan and his team are delivering this.  

“One of our past innovations was to adjust the field of view of the camera with intelligent zoom,” he explains. “So, if you don’t need the 180 degrees because you are alone, the camera will detect where you are and frame you.

“We wanted to build on this so worked with movie makers and directors and asked them about the psychology of human perception and how people absorb information. So, for example, if two people are sat next to each other on camera, other solutions would focus on the person speaking and potentially cut the other person out, or flip between the speakers, which can be annoying.” 

Jabra combined its beamforming technology and precise audio detection with facial recognition to understand who is speaking and present that to other members of the conversation.  

“It may not seem like a big deal, but your brain registers all the extra information and it can become distracting,” says Khan. “We fundamentally believe that powerful edge intelligence enables superb experiences. We are working hand-in-glove with the leading providers such as Microsoft because they really value this incredibly powerful AI being built into a device.” 

Jabra’s products can improve business processes by facilitating more effective communication, preventing the fatigue caused by problematic conferencing solutions and enhancing collaboration among users. But Khan believes his solutions could also have an important place in education.  

“Panacast 20 is a more mobile version of Panacast 50 with a 105-degree field of view,” says Khan.  

The AI involved in Panacast 20 can offer light optimisation, which automatically adjusts the lighting around the speaker; intelligent zoom, which keeps the speaker centred as they move; and picture-in-picture, which can focus on something other than speaker, while still capturing them in a smaller image.

“Over the last year, many educators have been using our products and they love that they are untethered,” says Khan. “Great teachers are almost dramatic, they use voice, intonation, movement, gestures, pacing, all kinds of things to kind of hold our attention. With our device, they can continue doing all this while teaching remotely.” 

This article was originally published in the Summer 2021 issue of The Record. To get future issues delivered directly to your inbox, sign up for a free subscription.

Number of views (1494)/Comments (-)

Comments are only visible to subscribers.

Theme picker