Microsoft Thinks It Should Be Easier To Interrupt People During Meetings, And It’s Kind of Brilliant

If you’ve spent much time on video meetings over the last two years, you probably have mixed feelings about them. On the one hand, they have made it possible to stay connected during a time when it just wasn’t possible to be together in person. During the pandemic, video calls were how we did everything from work, to school, to yoga classes, and even family holidays. 

On the other hand, video calls are still mostly bad. 

Sure, they’re better than not being able to see and talk to each other, but they are a long way from being a how almost anyone wants to spend a significant amount of time. Even if you’re fully in favor of working remotely (as I am), you have to admit that video calls are not a good substitute for being together in person. 

Partly, that’s because people aren’t great at video calls. First of all, they’re exhausting. Second, they don’t really facilitate interaction unless you’re ruthless about keeping people engaged (which might explain why they’re so exhausting). Finally, even after two years, most people either haven’t quite figured out how to look and sound good on a video call, or they just don’t care anymore.

But, it’s also because software is never going to be able to authentically recreate the experience of being together in person. Despite the effort by tech companies, video calls aren’t much better than they were the first time someone sent you a link to join their Zoom room.

That’s not to say they aren’t trying. In fact, earlier this month, Microsoft announced two new features in Teams aimed at solving a few of the worst things about trying to talk to someone through a webcam and speaker.

First, the company is using some advanced artificial intelligence (AI) to better cancel out background noise. That includes reverb and room echo, making you sound like you’re using a high-quality microphone, even if you aren’t. It even improves your sound when you’re not wearing headphones with a microphone.

The second, far more interesting feature is designed to make it easier to interrupt people, which, when you think about it, is kind of brilliant. Here’s why:

Interruptions are a real thing that happens, especially when people are in a room together. You might want to ask a question or clarify something someone said before they move on. In real life, that’s a natural part of having a conversation. Often, it doesn’t even feel like an interruption, it’s just how we talk. Conversations naturally flow back and forth between two people.

On a video meeting, however, it’s not good. Video meetings are fine when one person is presenting, and everyone else stays on mute and only talks when they’re sure the lane is wide open. If anyone else starts talking, it all falls apart.

Mostly, that’s because video calls aren’t good at handling crosstalk, where two people are talking at the same time because the software is trying to prevent feedback. On a video call, if one person is talking, the sound of their voice is coming out of the speaker of everyone else’s device. If one of those other participants starts talking, their microphone would pick up both their voice as well as the sound coming out of the speaker, creating an echo and a feedback loop. 

So, most video conferencing software essentially mutes your microphone unless you’re talking. When you start talking, the software has to recognize your voice and do all of the processing required to filter out any background sounds. That usually results in a small delay before anyone can hear what you’re saying. During that time, part of the conversation gets cut off, leading to the awkward “no, you go ahead.” 

Microsoft, on the other hand, says it’s using artificial intelligence to eliminate the echo and make it easier for multiple people to talk at the same time.  Microsoft says it used an AI model trained on “30,000 hours of speech samples to retain desired voices while suppressing unwanted audio signals resulting in more fluid dialogue.” Or, put another, it’s making video conversations more natural and more like real conversations. 

That’s a big deal. I’m not sure anyone wants to actually spend more time on video calls, but I think we can all agree anything that makes them better is a win for everyone. 

The opinions expressed here by columnists are their own, not those of

Source link

Leave a Reply

Your email address will not be published.