Microsoft Teams users can now create 3D avatars that can be used in meetings instead of being on camera or using a webcam. The feature was first announced two years ago, and is now available in public preview beginning from today.
In a blog post on Monday, Microsoft said the new avatar feature gives users additional options when online. “Avatars for Microsoft Teams offers an alternative to the current binary option of video or no video,” says Avery Salumbides, a product marketing manager at Microsoft. “Avatars for Teams gives you that much-needed camera break, while still allowing you to collaborate effectively.”
The new Teams avatars will animate based on your vocal cues alone. That means your avatar will still function normally even when you do not have a PC that has a webcam. While the preview is publicly available from today, the roll out is not expected until May according to The Verge.
About a year ago, Microsoft said it would introduce new AI-powered voice quality improvements that should eliminate or at least improve poor room acoustic, which makes it hard for calls to be heard.
The company announced that it will be using machine models to improve room acoustics. With this, you will no longer sound like you are in an empty room or hiding somewhere. In an interview with The Verge, Robert Aichner, a principal program manager for intelligent conversation and communications cloud at Microsoft, said: “While we have been trying our best with digital signal processing to do a really good job in Teams, we have now started using machine learning for the first time to build echo cancellation where you can truly reduce echo from all the different devices.”
Prior to the announcement, the feature had been in test for months, and was used to measure models in the real world to ensure Teams users are noticing the echo reduction and improvements in call quality. Microsoft used 30,000 hours of speech to help in training its models, and captured thousands of devices through crowd sourcing where users are paid to record their voice and payback audio from their devices.
Once Teams detects sound is bouncing or reverberating in a room leading to shallow audio, the model will also convert captured audio and process it to make it sound like participants are speaking into a close-range microphone instead of an echoey mess.