Using Teams as a transcription device

Taylor, Earl, COR 0 Reputation points
2026-11-04T15:40:00.5166667+00:00

Good morning. The County runs a 6 month inpatient drug treatment program. We will be getting a client who is deaf, but doesn't know sign language. We will individual treatment sessions and group treatment sessions. For the group sessions, does TEAMS have a product that will transcribe what is being said by each of the 6-8 people in small group sessions? Will each individual speaker be identified so the deaf client can follow along?

any assistance is appreciated.

Microsoft Teams | Microsoft Teams for business | Other

2 answers

Sort by: Newest
  1. Lin V 1,965 Reputation points Independent Advisor
    2026-11-04T16:33:46.9433333+00:00

    Hi Earl,

    Microsoft Teams includes built-in Live Captions and Transcription features that may help in this scenario.

    For the best experience with 6-8 participants, it is recommended that each participant join the Teams meeting using their own Teams account and device whenever possible. Teams is generally better able to identify individual speakers when participants join separately. In addition, participants should speak clearly and avoid talking over one another.


    1/ About live captions

    If the goal is to help the participant follow the discussion in real time without necessarily saving a transcript, captions can be enabled during the meeting.

    If your users are using a business account:

    First, open Teams > ellipses next to your profile > Settings > General > turn New meeting controls ON.

    Then during the meeting:

    • More (ellipses) > Captions

    For reference: Use live captions in Microsoft Teams meetings

    In addition, each speaker should make sure that Automatically identify me in live captions and live transcripts is enabled:

    Teams > Settings > Accessibility > Automatically identify me in live captions and live transcripts

    If your users are using personal accounts, please refer to: Live captions in Microsoft Teams for personal use


    2/ Transcription

    If a written transcript is needed, the meeting organizer can start transcription during the meeting:

    • More (...) > Record and transcribe > Start transcription
    • or More (ellipses) > Transcript (If new New meeting controls is enabled)

    Microsoft's transcription feature can display the transcript in real time and may identify individual speakers when they can be recognized.

    Please note that if all 6-8 participants are speaking in the same room through a shared microphone, speaker identification may be less accurate. For the best speaker identification experience, it is recommended that participants join the meeting using their own Teams accounts and devices whenever possible.

    For reference: Start, stop, and download live transcripts in Microsoft Teams meetings

    I hope this information is helpful. Let me know if you have any questions.

    Was this answer helpful?

    1 person found this answer helpful.
    0 comments No comments

  2. AI answer

    2026-11-04T15:41:51.0366667+00:00

    Use Teams live captions for accessibility without retaining a written treatment record; if each person must be named while 6–8 people share one room microphone, configure Intelligent Speaker recognition and have every speaker enroll a voice profile.

    What Teams can provide

    • Live captions display speech in real time with speaker attribution but are not saved.
    • Live transcription displays a real-time transcript and saves it afterward with timestamps and speaker attribution in OneDrive or SharePoint. ^1^
    • During a meeting, select More actions → Language and speech → Turn on live captions for captions, or More actions → Transcribe, select the spoken language, and select Confirm for a saved transcript. The exact available wording can vary by Teams client. ^2^

    For treatment sessions, captions are the lower-retention choice when no saved record is required. Confirm the County’s consent, health-record, and retention requirements before enabling transcription. Teams can also require participant agreement before recording or transcription. ^3^

    Identifying 6–8 people sharing one room

    A normal shared microphone may attribute speech only to the room. To identify individuals by name, use either:

    • Teams Rooms: Teams Rooms on Windows or Android with a Teams Rooms Pro license.
    • BYOD room: A current Teams desktop client on Windows or macOS, with the room host licensed for Teams Premium or Copilot.

    Six to eight participants are within Microsoft’s recommended maximum of 10 in-person attendees for best recognition precision. For either option:

    1. The organizer’s mailbox must be in Exchange Online. Identification is unsupported with on-premises Exchange or third-party calendars.
    2. Schedule a Teams meeting and invite each speaker individually.
    3. Each person who should appear by name must use an appropriate organizational Teams identity and create a voice profile in desktop Teams under Settings → Recognition.
    4. Where available, each speaker should also check Settings → Captions and transcripts → Automatically identify me in meeting captions and transcripts.
    5. Enable transcription and speaker-recognition policies.
    6. Have the deaf client join as an authenticated participant; anonymous participants cannot view the live transcript. ^2^

    These requirements and licensing differences are documented in the Intelligent Speaker guidance. ^4^

    Tasks for the County’s Teams administrator

    Enable Transcription and permit Live captions under Teams admin center → Meetings → Meeting policies. Both the organizer and the person starting transcription must be covered by a policy that permits transcription. ^1^

    For a Teams Room, apply Attribute mode to the policy assigned to the Teams Rooms account. For BYOD, enable speaker attribution:

    # Permit transcription and named speaker attribution.
    Set-CsTeamsMeetingPolicy -Identity "<PolicyName>" `
      -AllowTranscription $true `
      -RoomAttributeUserOverride Attribute
    
    # Required for speaker recognition in a BYOD meeting space.
    Set-CsTeamsAiPolicy -Identity Global -SpeakerAttributionForBYOD Enabled
    

    Policy changes can take up to 48 hours; signing out and back in can apply them sooner. Under Attribute mode, enrolled people appear by name while unenrolled people appear as Speaker. If displaying names is unnecessary, an administrator can use Distinguish instead, which separates voices as Speaker 1, Speaker 2, and so forth without names. ^4^

    Voice profiles require participant opt-in. People can choose not to be identified, and incorrect identification can be corrected from the meeting transcript. ^5^

    Run a complete trial in the actual treatment room before relying on it: test microphone placement, overlapping speech, the selected spoken language, the client’s view of both captions and the transcript pane, and every participant’s attribution. Automated captions and voice recognition can make errors and should not be treated as guaranteed accommodation accuracy; consider professional CART captioning if verified human accuracy is required.


    References

    1. Admins - Manage transcription and captions for Teams meetings
    2. Start, stop, and download live transcripts in Microsoft Teams meetings
    3. Manage Microsoft Teams meeting recording and transcription options for sensitive meetings
    4. Manage voice recognition for Intelligent Speaker
    5. Use Microsoft Teams Rooms to identify in-room participants in a meeting transcription
    AI-generated content may be incorrect. Read our transparency notes for more information.

    Was this answer helpful?

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.