In Teams, you have the option to do a ‘Voice and Face Enrollment’. But what is this feature? What’s in it for you? And what are the administrative implications? Let’s dive in!
What is Voice and Face Enrollment in Teams?
With this feature in Teams, you can enroll your voice and face. Sounds cool, right? It basically ’learns’ Teams how you sound and what you look like, so Teams can use it in your advantage. For example, it can help with background noise suppression, or even help you find your way in a meeting by recognizing your voice and face.
What’s in it for you?
By enrolling your voice and face in Teams, you can benefit from improved experiences both on a regular Teams client, and when joining a meeting from a Teams Room. A list of the benefits:
Noise suppression
Teams can use your voice profile to better suppress background noise, making your voice clearer to others in meetings. See it as a sort of noise-canceling headphone, but for your voice in Teams. The Teams service will use the profile of your voice to better distinguish it from background noise, and suppress that noise more effectively.
Meeting join experience
When you join a meeting from a Teams Room, the system can use your voice and face profile to recognize you. This can allow for a more personalized experience, such as automatically adjusting camera angles to focus on you, or even providing real-time transcription and translation services that are tailored to your voice. For example, if you have a strong accent, the system can use your voice profile to better understand you and provide more accurate transcriptions.
Speaker attribution in live captions
When you have live captions enabled in a meeting, Teams can use your voice profile to attribute spoken words to you. This means that when you speak, your name will appear next to the captions, making it easier for others to follow along and understand who is speaking. This can be especially helpful in larger meetings where multiple people are speaking, as it provides clarity and context to the conversation. Even when you join the meeting from a Teams Room, your voice profile can still be used for speaker attribution, ensuring that your contributions are recognized regardless of how you join the meeting. In stead of just showing ‘Speaker 1’, ‘Speaker 2’, etc., it will show your name, making the meeting experience more personalized and engaging.
This is even better when you combine Teams with Copilot, or the facilitator agent, as it can use the speaker attribution to provide more accurate summaries and action items based on who said what in the meeting. Imagine that the action items Facilitator picks up are automatically added to your to-do list, because Teams knows that you were the one who said ‘I will take care of that’. This can help you stay organized and ensure that important tasks are not forgotten after the meeting.
The voice profile can also help improve the meeting recap, as it can use the information about who said what to provide a more accurate summary of the meeting. For example, if you said something important during the meeting, the recap can highlight that and make sure it’s not missed.
But what about privacy and security?
Teams, or Microsoft, of course stores the ‘biometrics’ data that comes with the voice and face enrollment. Microsoft lists some actions it takes to protect this data.
Data handling
There are some things Microsoft does to protect your data. For example, the data is stored in the same region as your tenant, to ensure compliance with data sovereignty and privacy regulations.
Both voice and face data are encrypted at rest and in transit - and they are protected by Microsoft’s security and privacy policies and practices. Microsoft also states that they will not share the data with any third parties, except as required by law or to provide the service.
Data retention
So how long will your data stay with Microsoft? Well: It depends. If you decide to ‘unenroll’, you data is removed imediately. Even if the admin changes to policy to allow you to enroll, you always have the option to unenroll and have your data removed.
If you do not unenroll manually, but your account is deleted, the data will follow the account deletion process, which means that the data will be deleted after 30 days. Of course, if your company has specific processes in place, you account may exist for a period of time after you leave the company, and during that time, your data will also be retained.
If you don’t actually use the data, it is automatically removed after 1 year. A part of the data is on you endpoint device, to you use the Voice Isolation feature. This local copy has a signature that expires after 14 days - after that is replaced with a new download when you use the feature again.
Set up biometrics in Teams as an admin
As an administrator, you can controll the option to allow users to enroll their voice and face in Teams. You can do this through powershell, using the CSTeamsAIPolicy policy. Within this policy, the EnrollFace and EnrollVoice settings control the enrollment optins for end users. By default, both settings are enabled, you can disable one or both of them.
Examples
To set up voice and face enrollment for all users:
| |
To disable voice and face enrollment for specific users, you can create and assign custom meeting policies:
| |
Of course, the PowerShell cmdlets are well documented.
Teams Rooms people recognition
To use the people recognition feature in Teams Rooms, there are some additonal steps you need to take.
You can do this via the Teams Room Pro portal, or use PowerShell to change or create a Teams Meeting Policy:
| |
End user enrollment process
The enrollment process for end users is pretty straightforward. When the feature is enabled, users will see a prompt in the Teams client to enroll their voice and face. They can choose to enroll or skip the process. If they choose to enroll, they will be guided through the process of recording their voice and taking a picture of their face. The process is designed to be user-friendly and should only take a few minutes to complete. Once the enrollment is complete, users can start benefiting from the improved experiences in Teams meetings.
Wrap-up
The face and voice enrollment options in Teams can provide a more personalized (and engaging) meeting experience, with improved noise suppression, speaker attribution, and meeting recaps. As an admin, you have control over who can enroll and benefit from these features, and you can easily manage this through PowerShell. Just keep in mind the privacy and security implications of storing biometric data, and make sure to communicate this clearly to your users.
For end users, the feature is always opt-in: they need to enroll their voice and face themselves, and they can choose to unenroll at any time if they like.
Especially for organisations that work in noise environments, the improved noise suppression can be a game changer.
The speaker attribution can be a help - especially if you are looking into using Copilot or the Facilitator agent, as it can provide more accurate summaries and action items based on who said what in the meeting.