MICROSOFT TEAMS TRANSLATION

Translating a Microsoft Teams call, in both directions, without adding anything to Teams

VoxisLive runs beside Teams instead of inside it: it reads the audio your computer is already playing, speaks it back to you in your language, and — if you set up a virtual microphone — sends your own words into the call in theirs. Nothing is installed into Teams and nothing joins the meeting.

This page is the Teams-specific version of Meeting mode. The mechanism is identical for every conferencing app; what changes from one to the next is which menu holds the microphone and speaker dropdowns. Teams hides them in two different places depending on whether you are in a call yet, which is the main reason this page exists separately from the Zoom one.

What happens during the call

Meeting mode runs two independent translation sessions at the same time, one per direction. They do not share a connection, and neither waits for the other.

  • Incoming — the audio Teams is playing through your sound device is captured, translated, and spoken back to you in the language you chose. This half needs no extra driver and no change inside Teams.
  • Outgoing — the voice from your real microphone is translated, and the translated speech is played into a virtual microphone that you then select as your mic in Teams. This half needs one free driver installed first.

Both directions run a few seconds behind the speaker, the way a simultaneous interpreter does. That span is part of interpreting, not a slider you can turn down.

No bot, no tenant admin, no app in Teams

VoxisLive never connects to Teams. It captures at the operating-system audio layer — the same stream your speakers read from — so the application producing the sound is irrelevant to it. A Teams meeting, a Teams call, a channel meeting you are only listening to: to the capture, all of it is just audio Windows is playing.

For a Teams user specifically, that removes the three approval steps a translation tool normally runs into:

  • Nothing is added to the participant list. There is no bot to admit from the lobby, no extra attendee, and no notification for the organiser.
  • No tenant administrator has to approve anything. Nothing is published to your organisation's Teams app catalogue, and no Microsoft 365 consent prompt appears, because no application is being installed into Teams at all.
  • No calendar or account connection is requested. VoxisLive does not need to know that a meeting exists — you start the session yourself, when you want it.

If your organisation blocks third-party meeting bots or keeps the Teams app catalogue locked down, that policy is not something you are working around here. There is no app to allow and no bot to approve in the first place.

Incoming: hearing the other side

In Meeting mode the capture is pinned to the driverless path: it reads the system mix while excluding VoxisLive's own process, which is what keeps the translated voice out of the capture. Without it the app would hear its own output and translate the translation. That is not a setting — it is how meeting capture is built.

Because the incoming direction reads your sound output, it does not depend on your microphone at all. If the outgoing half is not set up, or stops working, you still hear the other side in your language.

Outgoing: being heard in their language

For the other participants to hear you in their own language, the translated voice has to get into Teams — and Windows only lets one application send audio into a call through a microphone device. So the outgoing direction needs a virtual microphone: a free driver (VB-CABLE is the common one) that presents itself to Windows as an ordinary input device while actually carrying whatever another application feeds into it.

VoxisLive plays your translated speech into that virtual device, and Teams listens to it as though it were a headset. Without a virtual microphone installed, Meeting mode runs listen-only: you keep hearing the other side in your language, and they keep hearing your original voice. Nothing breaks — one direction is simply absent.

The other side hears a synthetic voice speaking your words a few seconds after you speak. It is not made to sound like you, and the delay is audible by design.

Where the device settings live in Teams

Teams exposes the same two dropdowns in three places, and which one you reach depends on where you are:

  • Before a meeting — on the pre-join screen, the audio settings control next to the Join button opens speaker and microphone selection.
  • During a meeting — the More (…) menu in the meeting toolbar leads to settings and a device panel that opens beside the call.
  • Outside any call — your profile picture at the top right, then SettingsDevices, which also holds the test call Teams uses to confirm the devices work.

Microsoft has moved these controls between Teams versions more than once. If a menu is not where this page says, look for the word Devices — that is the page you want, whatever it is nested under this month.

  1. In Teams, set Speaker to your headphones. Not the virtual cable — see the warning below.
  2. In Teams, set Microphone to the virtual cable's input device if you want the outgoing direction. Leave it on your real microphone if listen-only is enough.
  3. In VoxisLive, pick the Meeting scenario and select your real microphone as its input — that is the voice it translates.
  4. Set the two languages: the one you want to hear, and the one the other side should hear.
  5. Start the session, then join the Teams meeting as you normally would.

Install the virtual cable before starting Teams. Teams enumerates audio devices when it launches, so a driver installed while it is running usually will not appear in the dropdowns until you quit and reopen the app.

Meeting minutes are paid minutes. The free tier does not cover Meeting mode — the daily free allowance applies to the one-way listening modes only. Every minute of a translated meeting, in either direction, is billed against a paid balance. Check pricing before you schedule a long call, so a meeting does not end mid-sentence.

Teams desktop app or Teams in a browser — the same either way

Because capture happens at the operating-system audio layer rather than inside the application, the incoming direction does not care whether Teams is the installed Windows client or a tab in Chrome or Edge. Both are just applications playing sound, and both are captured identically.

The outgoing direction differs only in how you select the microphone. The desktop client reads its devices from Windows directly; a browser tab asks the browser first, so the site needs microphone permission and the virtual cable has to be selectable in the tab. Once it is selected, the same translated audio reaches the meeting either way.

When something sounds wrong

The confusing failure is a routing one. If you set Teams' Speaker to the same virtual cable that VoxisLive is feeding your translated voice into, that cable becomes both the way sound leaves Teams and the way sound enters it. Call audio and the translated voice circle back into the meeting, and the other participants hear what sounds like a third person on the line, half a beat behind everyone else.

  • The virtual cable belongs in one Teams dropdown only: Microphone.
  • Teams' Speaker should always be a real output device — your headphones.
  • If you hear yourself, or someone reports an extra voice, check those two dropdowns before anything else.

What this is not

  • It is not a Teams integration. It does not read the participant list, chat, meeting title, calendar entry or recording, and knows nothing about the call beyond the sound it produces.
  • It does not remove interpreting delay. Both directions run a few seconds behind live speech.
  • It does not make the outgoing voice sound like you — the other side hears a synthetic voice speaking your words.
  • It is not covered by the free tier. Meetings run on a paid balance; see the note above.
  • It does not replace a human interpreter where one is legally or contractually required.

Where to go next

For the general explanation of two-way meeting translation — how the two language pickers work and why neither one asks what language you speak — start at the meetings hub. The same setup covers Zoom and Google Meet with their own device menus. If a device is missing from a dropdown, or you are not sure which output Windows is actually using, work through the audio setup checklist — most Meeting-mode problems turn out to be one dropdown pointing at the wrong device. Known issues and their fixes are in troubleshooting.

FAQ

Common questions

01Does my IT department have to approve or install anything in Teams?
No application is installed into Teams, so there is nothing to publish to your organisation's app catalogue and no tenant administrator consent to obtain. VoxisLive is a Windows application that captures the audio your computer is already playing; Teams is never asked for anything. Installing software on a managed workstation may still be governed by your own IT policy — that part is between you and your administrators, not between us and Teams.
02Will the other participants see that I am using it?
Nothing joins the meeting, so nothing appears in the participant list and no Teams notification fires. What is noticeable is the outgoing direction itself: if you enable it, they hear a synthetic voice speaking your words a few seconds after you speak, rather than your own voice. That is audible by design.
03Do I need a virtual microphone?
Only for the outgoing direction. Hearing the other side in your language works with no extra driver at all. Without a virtual microphone, Meeting mode runs listen-only: you are translated to, but not translated for.
04Can I use my free daily minutes for a Teams meeting?
No. The free tier does not cover Meeting mode. Meetings run on a paid balance, and both directions belong to the same session.
05Does it work with Teams in the browser, or only the desktop app?
Both. Capture happens at the operating-system audio layer, so the application producing the sound does not matter — a Teams tab in Chrome or Edge is captured exactly like the desktop client. For the outgoing direction you additionally have to be able to select the virtual microphone where the call is running, which in a browser means granting the tab microphone permission first.
06Someone in the call says there is an extra voice echoing. What is wrong?
Almost always Teams' Speaker is pointed at the virtual cable. The cable then carries sound in both directions and the call feeds itself. Set Speaker back to your headphones and leave the cable in the Microphone slot only.