Google Meet expands speech translation to 63 languages, with a choice for each listener

On October 22, 2026, Google announced expanded Meet speech translation. It supports 63 languages, detecting the spoken language and translating it into each listener's choice. Access is rolling out on web and mobile; Meet hardware currently supports listening only.

Instead of setting one language pair for the entire meeting, participants enable translation and choose a listening language individually. The system also attempts to detect when a speaker changes languages. Rollout starts October 22 for Rapid Release organizations and November 2 for Scheduled Release organizations, so the control may not be visible yet.

Consider a group with Chinese, English and Spanish speakers. The previous setup required one translation language pair for the whole meeting; the new approach lets each listener choose. This scenario explains the use, rather than reporting a translation quality test.

Audio and text can complement each other. With both speech translation and captions enabled, the captions match the language of the translated audio. Ask speakers to repeat names, dates or specialized terms and confirm them in writing. The user guide notes a delay of a few seconds and possible detection or translation errors. Leaving time for others to hear the translation helps a discussion proceed more smoothly than talking over it.

Scheduled meetings require a host with an eligible plan; in a direct one-to-one call, at least one participant must qualify. Supported plans include Workspace Business Standard/Plus, Enterprise Standard/Plus, Frontline Plus and Google AI Pro/Ultra, along with the education and add-on licenses in the announcement. Organization settings must allow it, and age and regional restrictions also apply.

1. Join with a Meet client that has received the feature. On desktop, hover over the captions and translation control at the bottom and select Translate speech. It starts off at the beginning of a meeting.
2. On first use, select the language you want to hear and choose Next. Select Translate me if you agree to let others hear your translated speech. Each participant enables the feature individually; you do not make that choice for everyone.
3. For text support, choose Turn on captions from the same control. You could first say, “Our next discussion is on Wednesday afternoon,” and check what the other person hears and reads. This is a suggested short test, not a guarantee of accuracy.
4. Watch the translation status icon. To stop listening to translations, select the Translating to button for your language. To stop translating your own voice, choose Don't translate me in speech translation settings.

Eligible base plans and Google AI Pro allow up to 90 minutes per meeting; the higher access licenses listed in the announcement and Google AI Ultra allow up to eight hours. Translated speech is not provided in live streams or meeting recordings. Client-side encrypted meetings, dial-in or third-party interoperability calls, and adaptive audio are also unsupported. Check text separately before using it as a formal record; live translated audio does not mean a recording will preserve it.