Google Meet Language Interpretation: Your Options

Google’s Meet Help doesn’t describe separate interpreter audio channels of the kind Zoom offers. What it describes are machine features: live captions, translated captions and Speech Translation, with availability and limits that differ by feature. For simultaneous interpretation in a Google Meet call you have three routes: use Google’s built-in features where your plan and languages fit, bring a human interpreter into the call, or add a live AI translator such as MeetVoice, which shows subtitles and reads the translation aloud for any pair of 32 languages.

What Google Meet Offers Out of the Box

Google’s Meet Help describes three language features — live captions, translated captions and Speech Translation — and no separate interpreter audio channel.

Live captions transcribe speech in the language it is spoken. They are shown only to the person who turns them on and don’t translate anything.

Translated captions show the speech as translated text. Google lists them for these Google Workspace editions — Business Standard, Business Plus, Enterprise Standard, Enterprise Plus and Google AI Pro for Education; Business Starter and personal accounts are not on the list. They work in Meet on a computer and in the Meet mobile apps, and Japanese, Chinese and many other languages are supported. It is still text: nobody hears a translation.

Speech Translation is the closest thing to interpretation that Google ships. It translates a speaker’s voice in near real time, in a voice like theirs. Google announced general availability for businesses on 4 February 2026. Current Help lists these limits:

• English paired with French, German, Hindi, Italian, Portuguese or Spanish — there is no Japanese, Chinese, Russian or Polish, and no pair without English;
• one language pair per meeting;
• a delay of a few seconds, and a 90-minute limit per meeting;
• it is turned on from a computer; meeting room devices can only listen;
• eligible plans include Business Standard and Plus, Enterprise Standard and Plus, Frontline Plus and Google AI Pro and Ultra, and admins can turn it off;
• each speaker must allow their voice to be translated;
• it is not available in recordings or live streams.

If your meeting is English–Spanish, organized on an eligible plan and shorter than 90 minutes, Speech Translation may be all you need. The comparison with Google Meet captions goes through the differences in more detail.

Sources: Speech Translation · Translated captions · Google Workspace Updates

Bringing a Human Interpreter into Google Meet

A professional interpreter is still the best choice for negotiations, legal or medical conversations and public events, where nuance matters more than cost. When an interpreter joins as a regular participant and uses the same meeting audio as everyone else, participants hear both the speaker and the interpreter.

That works for consecutive interpretation: the speaker pauses and the interpreter repeats what was said in the other language. It roughly doubles the length of the meeting, but it needs nothing beyond Meet itself.

Simultaneous interpretation, where the interpreter translates while the original speaker keeps talking, is harder: in a single call both voices mix. Teams work around this with a second call or phone line that only the listeners of one language join. This can work for a planned event, but it requires booking an interpreter in advance and a separate audio connection — rarely practical for a daily stand-up.

Live AI Interpretation with MeetVoice

MeetVoice adds simultaneous interpretation to the Google Meet calls you already have, without an interpreter and without anything to install for the other participants.

• Subtitles and voice. Incoming speech appears as subtitles in your language and, by default, is read aloud as a translated voice while the original audio is turned down. The delay is measured in seconds and varies with speech and connection.
• Your side, if you want it. Translation of your own speech is off by default. Turn it on in the extension to send your speech to the call as a translated voice in one selected language; it doesn’t create a separate language channel for each listener.
• Any pair of 32 languages. English–Japanese, German–Russian, Polish–Chinese: pairs don’t have to include English. If more than two languages are spoken, multilingual mode translates incoming speech in 2–4 expected languages, chosen from a set of ten, into yours; translation of your own speech needs the regular language-pair mode.
• Any Google account. MeetVoice captures the audio of your Google Meet tab on your own computer rather than joining the call as a bot, so it works the same with personal accounts and every Workspace edition, Business Starter included.
• Only you see the subtitles. They are drawn in your own Chrome tab, so other participants don’t see them — unless you share that tab or your screen.

How to Set Up Simultaneous Interpretation in Google Meet

Install and open the MeetVoice desktop app, then add the Chrome extension. Choose your language and the other participants’ language in the extension, join the Google Meet call and start translation. Translation of your own speech is optional and is turned on separately.

It needs Google Chrome and the MeetVoice desktop app for Windows, or for macOS 13 or later on Apple Silicon. The desktop app runs the translation server on your computer; the Chrome extension shows the subtitles in your Meet tab. The quick start guide walks through the setup, and how live translation works in Google Meet explains the rest.

Which Option Fits Your Meeting

	Human interpreter	Google Speech Translation	MeetVoice
What you get	Spoken interpretation	Translated voice	Subtitles and translated voice
Languages	Whatever your interpreter speaks	English with 6 languages	Any pair of 32 languages
Account	Any	Eligible plans only	Any Google account
Time	Booked per session	Up to 90 minutes per meeting	Hours included in your monthly plan
Cost	Per session, booked in advance	Included in eligible plans	30-minute free trial, then a monthly plan
Best for	High-stakes events	English-centred meetings on eligible plans	Everyday calls in supported language pairs

Long meetings with MeetVoice draw on your plan’s monthly hours. Speech recognition reconnects automatically about every two hours, which can briefly interrupt the translation, and if MeetVoice recognizes no speech for 15 minutes, translation stops until you start it again. Your own speech keeps the session active only when translation of your speech is on.

For a board meeting or a court hearing, book a professional interpreter. For the weekly sync with a team in Tokyo, São Paulo or Warsaw, start MeetVoice translation when the call begins — no one to book, nothing for the other side to install.

MeetVoice has a 30-minute free trial, no card required: download the desktop app from the download page, add the Chrome extension and try it on your next cross-language call.