Posey's Tips & Tricks

Microsoft Teams Gets Much-Needed Audio and Video Quality Improvements

Say goodbye to annoying echoes during your next Teams meeting.

Ever since the onset of the pandemic over two years ago, online meetings have become a staple of everyday life. In some ways these video calls have become almost a parody of themselves due to certain annoyances that happen way too often.

Even though a lot of these annoyances stem from attendees with bad habits (and I may be one of the worst offenders), other annoyances are less about the attendees and more about the online experience. After all, we've probably all been on video calls in which someone's audio configuration caused a severe echo. I've also seen plenty of instances in which audio drops for no reason or where two people try to speak at the same time with completely unpredictable results. Thankfully, Microsoft seems to be taking steps to do something about these types of issues.

Over the last few years, Microsoft has invested heavily in AI-based speech recognition. In doing so, the company added a transcription feature to Microsoft Office and has even bolstered Office's dictation capabilities through the purchase of Nuance -- maker of Dragon Naturally Speaking.

Microsoft's newfound interest in speech recognition was hardly surprising. After all, good speech recognition capabilities are necessary for transcribing Teams meetings. They are also necessary for generating real-time close captioning. In the back of my mind though, I always wondered if speech recognition data might be useful for applications other than dictation and transcription.

While Microsoft has not indicated that it speech recognition capabilities have been used in the audio improvements that the company is introducing for Teams, Microsoft has confirmed that these improvements are based on the use of AI and that over 30,000 hours of speech samples were used to train the company's new machine learning model. It seems likely that Microsoft may very well have leveraged some of its speech recognition assets as a tool for improving Teams audio quality.

So what are these Teams audio improvements anyway? The biggest, and arguably most significant improvement that Microsoft is introducing is echo cancellation. Echoes generally occur when a meeting participant places their microphone too close to the speaker that is producing meeting audio. That audio gets picked up by the microphone, resulting in acoustical feedback. This problem is especially common when a meeting participant uses a laptop without a headset (leveraging the device's built-in speakers and microphone).

The audio training that Microsoft is done for Teams will allow Teams to tell the difference between audio that is spoken directly into a microphone and audio that is coming from a speaker that has been placed in close proximity to a microphone. Being able to differentiate between the two will allow teams to filter out audio that would ordinarily result in an echo.

Teams will also be able to filter out echoes that are caused by sources other than acoustical feedback. Sometimes when meeting participants are in large rooms (such as a conference room) sound may reverberate off of the walls and other surfaces, resulting in another type of echo. Teams is being designed to filter out these types of echoes as well.

In addition to helping to remove echoes, Microsoft is also working to suppress background noise. This will be a welcome improvement for anyone who has had to endure Teams meetings in which one of the participants had a crying baby, a barking dog, or a television blaring in the background.

One last improvement that Microsoft is making to Teams audio is a change that will make it less disruptive when two meeting participants accidentally speak at the same time. Because Teams is designed with full-duplex audio, participants can speak and here at the same time. Unfortunately that means that if two meeting participants talk at the same time, and one of those participants has a speaker that is too close to the microphone, it can cause a disruptive echo for everyone. Thankfully, Microsoft is figured out a way of filtering that out too.

It remains to be seen what these audio improvements will actually sound like, but I have high hopes for them. In addition to the audio improvements, Microsoft is also going to be taking steps to optimize video. The company is using AI to deal with bandwidth constraints and Teams is going to be outfitted with a feature that will automatically adjust video brightness and focus in a way that should deliver the best possible picture quality.

I applaud Microsoft for taking the time to figure out ways of improving audio and video quality in Microsoft Teams. I tend to think of these improvements as "quality of life" updates because unlike many other feature updates, they have the potential to greatly improve the overall experience of using Microsoft Teams.

About the Author

Brien Posey is a 21-time Microsoft MVP with decades of IT experience. As a freelance writer, Posey has written thousands of articles and contributed to several dozen books on a wide variety of IT topics. Prior to going freelance, Posey was a CIO for a national chain of hospitals and health care facilities. He has also served as a network administrator for some of the country's largest insurance companies and for the Department of Defense at Fort Knox. In addition to his continued work in IT, Posey has spent the last several years actively training as a commercial scientist-astronaut candidate in preparation to fly on a mission to study polar mesospheric clouds from space. You can follow his spaceflight training on his Web site.


comments powered by Disqus