When AI shows up at your remote meetings

With the rise of remote working, what will our meetings look like in the future? This was the question asked by a team in the Orange Smart Working project. Presented exclusively at the 2019 Research Exhibition, the “Future Meeting Experience” demo simulates a meeting in which attendees’ actions and interactions are enriched by artificial intelligence, even when meeting remotely.

“The attendees no longer need to seek out the information themselves. Instead, the information comes to them.”

Attending a remote meeting – whether you are one of those physically present or not – can be frustrating. Who is talking? How are the people in the room interacting? Most often, misunderstanding and dissatisfaction ensues, and the remote attendee can feel sidelined. In view of these constraints, a team from the Orange Smart Working research project, led by Cédric Floury, has looked into the problem. The idea Is to offer the virtual attendee a meeting experience that is just as real as if they were in the room. And the team has successfully stepped up to the challenge, presenting their “Future Meeting Experience” at the 2019 Orange Research Exhibition.

Content relayed in real time

For their demo, the team has simulated an HR meeting taking place in a normal room. The attendees are sitting around a conference table. Each person has a stand in front of them holding a personal smartphone, allowing the microphone and camera to capture what is said during the meeting. During the session, the name of a company employee is mentioned. Immediately, the latter’s job description appears on the attendees’ phone screens, whether they are physically present or not.

How does it work? “We used artificial intelligence (AI) to relay content in real time to the phones. The attendees no longer need to seek out the information themselves. Instead, the information comes to them. The solution is proving even more useful for people who are remote”, says Cédric Floury, the project leader.

In another example, one of the attendees refers to an employee’s eligibility for professional training and mentions the acronym CPF (“Compte Personnel de Formation” or “Personal Training Account”). Instantly, all the information related to this training and even the relevant article of the corresponding law appears on a dedicated web page.

“It’s all down to machine learning algorithms. The machine is given a data string, in this case the French Labour Code, and it will then create mathematical links to suggest appropriate content. It even applies a confidence score to the result”, Cédric Floury adds.

360° videoconferencing

Finally, so as not to lose anything of the meeting, a mind map is created automatically as the conversation is taking place. This is a diagram on which all the ideas contributed in the meeting are visually represented. “It makes it possible to review key moments – a significant advantage for remote attendees”, explains Floury. “As well as this, remotees can experience an even greater feeling of being present thanks to the videoconferencing application called “Future visio experience”, which is still under development. All that is needed is for the people in the room to connect up to a 360° camera. And with the upcoming arrival of 5G, the experience promises to be even better!

Read also on Hello Future

NORIA: Network anomaly detection using knowledge graphs

GettyImages - machine learning - research recherche

Machine-Learning-Based Early Decision-Making (ML-EDM)


AI in Medicine: Uses and Consequences on Radiology Work


Green and Local Energy Exchanges and Optimizing Network Consumption


Advocating for ethical and responsible AI by design


Orange 2021 Research Exhibition: let’s prepare for the future!


Simulated data on household activity: we evaluate how realistic it really is!


Measuring network usage for each mobile application