Google Meet extends real-time speech translation to 63 languages, on the web and on Android and iOS. The biggest change is that the platform automatically detects the language each participant speaks, without requiring a single language pair to be set for the whole meeting, as was the case before.
The feature is powered by Gemini and, according to Google, uses the Gemini 3.5 Live Translate model. Each participant can turn translation on or off for themselves and choose the language they listen in. Languages a user already understands can be registered in their profile, so they are not translated. Captions adapt to the translated audio as well.
The system is designed for groups with members speaking different languages, so everyone can follow in their own language without everyone having to switch to the same one. Google says audio recordings are not stored and users' voices are not used to train the model.
Who can use it
Translation is not available to every account. The meeting organizer must have an eligible plan: the sources list Google AI Pro and Ultra, several Workspace Business Standard and Plus, Enterprise Standard and Plus, Frontline Plus and Google AI Pro for Education plans, along with some AI add-ons. For scheduled meetings, the organizer needs an eligible plan, while external guests can still take part. In direct one-to-one calls, it is enough that one participant has an eligible plan.
Free accounts and Business Starter are not among the eligible plans. The sources also mention monthly usage limits that vary by plan, an exclusion for participants under 18, and restrictions tied to region and administrator settings.
The sources note that translation is not instantaneous: words arrive a few seconds late. It is not a human interpreter.
The list includes languages such as Italian, English, Chinese, Japanese, French, Spanish, Arabic and Hindi, along with regional variants such as Cantonese. Sign language support is not part of this announcement.
No comments yet. Be the first!