How to translate a meeting in real time

Check what your conferencing platform actually does first — most translate to captions or a single language pair. If that fits, use it. If you need several languages, translated audio, or external guests, you will need something alongside it.

Start with what you already pay for, because a surprising share of cases are covered and the rest fail for reasons worth knowing before you buy anything.

What the native platforms do

The four questions that decide it

1. Do people need to hear it or read it? Zoom produces no translated audio at all. If your participants are listening rather than reading, Zoom’s feature is not a partial fit — it is a different product.

2. How many languages at once? Google Meet supports one language pair per meeting. A call with Italian, Spanish and German participants is not possible on Meet at any price. This is the limit that catches most teams out, because it is architectural rather than commercial.

3. Are there external participants? Teams Interpreter excludes guests and anonymous users, and requires every participant to hold a Microsoft 365 Copilot licence — roughly three times the cost most buyers assume, since Teams Premium is not sufficient. For internal meetings this is workable. For client, supplier or candidate calls it usually is not.

4. Does the recording need to keep the translation? Microsoft’s documentation states that Teams recordings capture only the original meeting audio and do not include interpretation audio, and translated transcripts disappear when the meeting ends. Google Meet does not translate recordings or live streams either.

For a distributed team where most people watch afterwards, this removes most of the value.

When the native option is genuinely enough

Two languages, one of them English, all participants internal and licensed, under 90 minutes, and the recording does not need translation.

That is a real and common situation. If it describes you, use what you have and stop reading.

When you need something else

• Three or more languages in the same meeting
• Translated audio rather than captions
• External participants without your corporate licences
• Sessions over 90 minutes
• Recordings or transcripts that must keep the translation
• Ad hoc meetings, which Teams Interpreter does not support

Whatever you use, do these three things

1. Load a glossary — product names, acronyms, domain terms. Highest-leverage single step, in every tool that supports it.
2. Brief participants to pause at clause boundaries. Not mid-sentence. A trailing sentence gives the translation nothing to close on.
3. Put numbers, dates and names in the chat. Highest consequence, cheapest to make safe.

Test before you commit

Run one real meeting, with real speakers, not a scripted demo. Read speech and spoken speech stress these systems very differently, and a vendor demo is always the former.