Transforming the capabilities of Microsoft Teams through AI and machine learning

As many businesses return to the office, we continue to see significant growth in the use of online meetings and conference calls. Despite these communication & collaboration tools helping businesses to continue to operate during the COVID-19 pandemic, a whole host of opportunities have been identified to help transform the overall meeting and call quality. 

It has been widely reported that issues such as poor room acoustics and choppy video have been limiting the effectiveness of both online calls and meetings. Through the use of AI and machine learning, Microsoft has been able to identify and deliver some useful improvements for Microsoft Teams. 

In our latest blog article, we will take you through some of these improvements in Teams in more detail. 

Voice quality improvements 

Echo cancellation 

Over the course of calls & meetings, when an attendee is using their microphone too close to their speaker, it can often be the case where the sound loops between input and output devices which forms a very irritating echo. This has now been resolved with Teams making use of AI to identify the difference between sound from a speaker and the user’s voice, reducing the echo without impacting on speech or stopping multiple people speaking at the same time. 

Adjustments made for poor room acoustics 

In certain situations, room acoustics can lead to sound bouncing which can lead to the user’s voice sounding shallow. This is now being resolved as Microsoft Teams again through machine learning will be able to convert captured audio signal to sound as if users are close to the microphone. 

Background noise suppression 

There are no doubts that we’ve all had a meeting disrupted by a loud lorry, car alarm or animals/children making loads of noise. Microsoft has been working extremely hard over the years to rectify this issue. After being able to record significant improvements across key user metrics, Microsoft has put into action machine learning noise suppression as a default for Teams customers using Windows as well as those using Mac devices. 

Video quality improvements 

Game changing real-time screen optimisation adjusts for the content being shared 

It goes without saying that how presentations are received largely depends on how well an audience can read and watch shared videos. However, it is important to note that all kinds of shared content need varied approaches to ensure that the video quality is maintained particularly under internet streaming constraints. This has now changed with Teams using machine learning to easily detect and adjust the characteristics of the content in real-time, optimising the legibility of documents or smoothness of video playback. 

AI driven optimisation to keep videos looking fantastic 

We’ve all experienced some form of unexpected issues when it comes to network bandwidth which has led to a choppy video that quickly distracts people from your presentation. This is no longer a concern as Teams through AI driven optimisations can help to adjust playback in challenging even the most complex of bandwidth conditions, so any presenters will be rest assured that their video and screen sharing will be trouble free. 

Added brightness & focus filters that show you in the best light 

Despite the fact that you are never in full control of how the surrounding light is in your meetings, the new AI powered filters coming to Teams will provide you with the option to make adjustments to the brightness and a softer focus of all your meetings through a simple toggle in your device settings to help in low lighting situations. 

Next Steps? 

If you are a business that is looking to take advantage of the above new features or looking to implement Microsoft Teams for the first time, need additional Teams training or have any other Microsoft 365 related requirements then contact our IT experts today. 

Want to get the latest Microsoft news delivered into your inbox? This can be achieved if you have signed up to our mailing list.