A client speaks one language. Your team speaks another. Until recently, keeping a video call moving often meant stopping to interpret, switching between translation tools or relying on someone fluent to bridge the gap.
Google Meet’s real-time speech translation is designed to make those conversations easier. It listens as people speak and plays a translated version in each listener’s preferred language, with only a short delay. Google’s October 2026 update lists support for 63 languages on web and mobile.
For small and medium businesses, the promise is practical: fewer pauses in routine conversations with clients, suppliers and colleagues. But it’s not a substitute for a professional interpreter when the details really matter, and eligibility depends on your Workspace plan.
What live speech translation actually does
During a meeting, Meet detects a speaker’s language and translates their words into the listener’s selected language. It can also translate what you say for other participants, if you choose to allow it. The translated voice is designed to sound like the original speaker.
Each person can choose the language they want to hear, so participants don’t all need to use the same translation. The translation is delivered in near real time, but it isn’t instantaneous: expect a delay of a few seconds.
To hear a translation, participants need to enable the feature. Speakers also need to consent to having their voice translated. Google says the audio isn’t saved and models aren’t trained on your voice. See Google’s Speech Translation guide for current language, privacy and availability details.
Where it could help a business
The feature is most useful when people need to talk through everyday work without a shared language. For example:
- A sales or service team can have a first conversation with a prospective client.
- An office manager can coordinate a routine update with an overseas supplier.
- A project team can check progress with colleagues who prefer another language.
- A business owner can build rapport with a new contact without pausing after every sentence.
It may also help teams include more people in a discussion, rather than relying on one bilingual colleague to interpret for everyone. That colleague can still take part in the conversation instead of having to carry the whole meeting.
Think of it as a way to make routine conversations more accessible, not a guarantee that every word will land perfectly. For an important call, share an agenda in advance, allow time for clarification and follow up key decisions in writing.
If your business is improving how staff collaborate online, Cloud Computer Company can also help with cloud collaboration planning and user training.
How to turn it on
First, check that the meeting is covered by an eligible Workspace plan. For scheduled meetings, the host must have an eligible plan. In a direct Meet call, at least one participant must have one. If the host is eligible, other participants may also be able to use translation even if their own plan doesn’t include the feature.
In a meeting:
- Open Meet’s speech translation controls. On mobile, these are under More options and Captions & translation.
- Choose the language you want to hear.
- Turn on speech translation. If you want your own speech translated for others, enable that option and allow translation of your voice.
- Invite others to turn on the feature and opt in if they want their speech translated.
Your administrator can check that the Gemini setting for Meet is enabled in the Admin console. Google says this setting is on by default, but an administrator can change it. The setting alone does not make an ineligible plan eligible. The administrator setup guide explains the controls.
The feature is available on web and mobile, subject to plan, device and regional availability. Eligible business plans currently have a 90-minute limit per meeting. It isn’t available in Meet live streams or recordings. Participants using Meet hardware in a conference room can hear translations from people outside the room, but their own speech won’t be translated.
The honest limits
Google cautions that live translations can have more errors than translations made from recorded audio or text. Speech may be misheard, grammar may be wrong, and accents or poor network conditions can affect the result. Idioms and industry-specific terms can be particularly tricky. A translation may sound fluent while still missing an important detail.
There’s also a small delay. That’s manageable for a planned conversation, but it can make rapid back-and-forth, interruptions and jokes harder to follow. Google recommends waiting for the translation indicator to disappear before replying. If something doesn’t make sense, slow down, rephrase and confirm the meaning.
Use a qualified human interpreter for contract negotiations, legal, medical or safety discussions, sensitive employee matters, or any conversation where a translation error could have serious consequences. Live translation can help people get the gist and keep a routine call moving; it shouldn’t be the final authority on a critical decision.
Live conversation or translating a document?
These tools solve different problems. Meet’s speech translation is for spoken conversation as it happens. It lets people talk and listen in their chosen languages during a call.
Google Docs’ Tools → Translate document creates a separate translated copy of a document. That’s useful for reviewing a proposal, policy or set of instructions before or after a meeting. Asking Gemini to translate text in a document can help with a written task, too, but it won’t translate the live voices in a call or manage real-time turn-taking.
A simple workflow might use both: translate background material ahead of the call, use Meet translation to discuss it, then have a person review any written version that will be relied on or shared externally. Google’s document translation instructions explain how to create a translated copy.
What Business Starter and Standard users should check first
Business Starter does not include Meet speech translation as a plan feature. Business Standard does. Google’s Meet edition comparison lists speech translation for Business Standard and Business Plus, but not Business Starter.
That doesn’t necessarily mean a Starter user can never use translation. If an eligible account hosts a scheduled meeting, participants may be able to turn it on in that meeting. In a direct call, at least one participant needs an eligible plan. The host’s administrator settings, participant consent, device and availability can also affect access.
Before planning an important call around the feature, confirm your exact Workspace edition, check with your administrator and test it with the people and devices you’ll use. If your team is unsure how its plan and settings fit together, our Workspace consultancy can help you review the setup.
Make your next call easier to follow
Live speech translation could remove some friction from routine multilingual calls, but it works best with clear speech, realistic expectations and a human check when meaning matters. Check your plan first, try it in a low-risk meeting and agree on how your team will confirm important details.
Want help reviewing your Workspace setup or planning a smoother rollout? Book a free 15-minute consult or call 1800 312 972.
About Mathew
Mathew Hoffman started in IT in 1981 and held senior roles at State Bank of NSW, Minet Australia, Wilhelmsen Lines and Rothmans of Pall Mall. His experience also includes work connected with the Sydney 2000 Olympics. Since 2001, he has provided IT consultancy to small and medium businesses. Mathew was an original Google Partner in 2008, and the business re-branded as Cloud Computer Company in 2017. Based in Noosa, he enjoys cricket, which he has played and coached in Sydney and the Sunshine Coast, as well as family time, the beach and golf.
