Google has expanded speech translation in Google Meet to 63 languages, including Ukrainian. With Gemini 3.5 Live Translate, video meeting participants can hear translations of what others are saying with a delay of just a few seconds. The system automatically detects the language and attempts to preserve the speaker’s intonation and distinctive vocal characteristics. The update works on computers and smartphones, but availability depends on the organizer’s subscription plan and regional restrictions.
Key Takeaways
- On October 22, 2026, Google announced support for 63 languages in Google Meet speech translation, including Ukrainian.
- Gemini 3.5 Live Translate translates spoken language and synchronizes subtitles with the translation.
- Each participant can choose their own translation language, even if other participants are using different languages.
- The feature is available on computers, Android, and iOS, and is partially supported on Google Meet devices for conference rooms.
- Google AI Pro, Ultra, and select Google Workspace plans are supported. Guests do not necessarily need their own paid subscriptions.
- Translation is limited to 90 minutes or eight hours, depending on the plan.
What’s Changed in Google Meet
Speech translation is not an entirely new Google Meet feature. In early 2026, Google made it available on certain Workspace business plans, and in April began rolling it out on Android and iOS.
Previously, the system supported a limited set of language pairs. A single pair, such as English and Spanish, had to be selected in advance for the entire meeting.
The October 22, 2026 update changes that approach. Google Meet now supports translation between any combination of the 63 available languages, automatically detects the speaker’s language, and lets each listener choose their preferred translation language.
This means that participants in the same meeting can speak Ukrainian, English, Polish, and other supported languages, while each person hears the translation according to their own settings.
How Gemini Translates Speech, Not Just Text
The feature is powered by Gemini 3.5 Live Translate, Google’s audio model introduced in June 2026. It recognizes speech, translates it, and generates spoken audio in the target language.
Unlike conventional translated captions, users receive audio. Google says the model attempts to reproduce the original speaker’s intonation, pace, and pitch.
Translation happens almost in real time, but not instantly. Google warns of a delay of several seconds, which is needed to understand each phrase more accurately.
The model can handle switching between languages during a conversation. If someone starts a sentence in English and then switches to another supported language, Meet can automatically adjust its language recognition.
When translation and captions are enabled, the text displayed on screen matches the language of the spoken translation. At the same time, Google cautions that automatic language detection and captions may occasionally be delayed or contain errors.
Ukrainian Is Among the 63 Supported Languages
Ukrainian is explicitly listed among the officially supported languages for Google Meet Speech Translation.
Other available languages include English, Polish, German, French, Spanish, Italian, Portuguese, Japanese, Korean, Chinese, Turkish, and more.
It is important not to confuse the number of languages available in Google Meet with the capabilities of the underlying model. Google says Gemini 3.5 Live Translate supports more than 70 languages as a technology, but only 63 were available in Meet at the time of the announcement.
For Ukrainian companies, this creates a practical opportunity to hold international meetings in Ukrainian without necessarily requiring everyone to switch to English.
Who Can Use Google Meet Translation
The feature is not limited to Google AI Pro and Ultra subscriptions. Google has confirmed availability for several categories of users.
- Personal accounts: Google AI Pro and Google AI Ultra.
- Businesses: Google Workspace Business Standard and Business Plus.
- Enterprise customers: Enterprise Standard, Enterprise Plus, and Frontline Plus.
- Education: Google AI Pro for Education.
- Add-on packages: AI Expanded Access and AI Ultra Access.
There is an important benefit for teams on different plans: if a meeting is created by a user with an eligible plan, other participants, including external guests, can also use speech translation.
For a direct call between two people, it is enough for at least one participant to have an eligible plan.
This means a company does not necessarily need to purchase a separate AI subscription for every client or contractor it invites to a meeting. However, actual availability depends on the organizer, Workspace settings, and region.
How Long Translation Can Run
Google has set different time limits for using Speech Translation during a single meeting.
- Up to 90 minutes: eligible basic Google Workspace plans and Google AI Pro.
- Up to 8 hours: AI Expanded Access, AI Ultra Access, and Google AI Ultra.
These limits apply to the use of speech translation, not necessarily to the maximum length of the video meeting itself.
For short client calls, 90 minutes may be enough. For conferences, lengthy negotiations, and internal strategy sessions, the restrictions are worth considering during planning.
How to Enable Speech Translation
In the new version of Google Meet, participants can control translation individually. To do so, they need to open the relevant translation control during the meeting and choose the language in which they want to hear other participants.
The system also allows users to specify languages they already understand. Meet will then avoid translating speech in those languages unnecessarily.
Translation is turned off by default at the start of a meeting. Users decide for themselves whether to enable it and whether to allow their own voice to be translated for others.
For corporate accounts, an administrator can manage feature availability in Google Workspace settings. On smartphones, an up-to-date version of the Google Meet app is required.
Where the Feature Doesn’t Work Yet
Despite the expansion in language support, Google has introduced a number of technical restrictions.
- Speech translation is not available during Google Meet live streams.
- The feature does not work in meeting recordings.
- Meetings with client-side encryption, dial-in calls, and third-party interoperability systems are not supported.
- Translation is unavailable when Adaptive Audio is in use.
- Google Meet conference room devices can play translations of remote participants, but speech originating in the room is not translated for others.
Availability also depends on age and regional restrictions. As a result, even users on eligible plans may not see the feature if it has not yet been rolled out to their accounts.
What Google Says About Privacy
Google says users can independently prevent their own voice from being translated for other meeting participants.
According to the company’s official explanation, audio data is not stored for this feature, and the models are not trained on participants’ voices.
For business meetings involving confidential information, however, organizations should separately review their policies on AI features and sharing data with third-party services.
What This Means for International Teams
The most obvious use case is meetings with clients, partners, and contractors who speak different languages.
For example, a Ukrainian marketing team could present campaign results in Ukrainian while an international client listens to an English translation. The client could then reply in their own language without forcing everyone in the meeting to adopt a single shared language.
This could reduce language barriers during negotiations, but it does not eliminate the risk of inaccuracies. Advertising metric names, legal wording, financial terms, and specialized terminology still need to be checked.
The update also fits into Google’s broader strategy for voice-based AI. UAGEEK previously covered the Live Translate mode in Google Translate, while our Gemini overview brings together the key capabilities of the company’s AI ecosystem.
When the Update Will Reach All Users
Google began a rapid rollout of the update on October 22, 2026. For Scheduled Release domains, the gradual rollout started on November 2 and may take up to 15 days.
As of November 11, therefore, the feature may still be appearing in certain accounts. Google also warns that the availability of translation and synchronized captions may differ during the initial rollout.
Conclusion
The main change in Google Meet is not simply the increase to 63 languages, but the removal of a single language pair for the entire meeting. Each participant can now listen to translations in their preferred language, while the system automatically detects what languages others are speaking.
Support for Ukrainian makes the feature particularly useful for international teams from Ukraine. However, before using it in important negotiations, teams should check their plan, time limits, and the accuracy of translations of professional terminology.
Sources for Verification
Sources for Verification
Google Workspace Updates, October 22, 2026. Official announcement of 63 supported languages, automatic language detection, individual settings, subscription plans, time limits, and rollout schedule.
Expansion of Speech Translation in Google Meet
Google Meet Help — Use Speech Translation. Official list of supported languages, including Ukrainian, translation settings, user consent, and feature restrictions.
Google Meet Speech Translation Guide
Google Blog, June 9, 2026. Technical details of Gemini 3.5 Live Translate, support for more than 70 languages at the model level, and reproduction of voice intonation, pace, and pitch.
How Gemini 3.5 Live Translate Works
Google Workspace Updates, April 8, 2026. Earlier rollout of speech translation in the Google Meet mobile apps and the language-pair restrictions in place at the time.
Google Meet Translation Launches on Android and iOS
