Introducing Gemini 3.5 Flash Live Translate, our real time speech to speech translation model which supports more than 70 languages (both in and out), and is so natural.

It is available in the Gemini API, AI Studio, & Google Translate right now + coming soon to Google Meet!!

Gemini 3.5 Live Translate! We just shipped a real-time babel fish.

- 70+ languages, 2,000+ language pairs. - Natural translated speech, works in noisy environments. - Stays in sync with the speaker, no lag, no awkward pauses. - Auto-detects the language being spoken.

Available today in Google Translate (Android & iOS), the Gemini API (Public Preview), and Google Meet (Private Preview).

I genuinely think this is the beginning of the end of language barriers. Anyone can now speak and understand anyone.

3.5 Live Translate can convert speech into over 70 languages and processes it as it’s streamed - while keeping tone, pace and pitch intact - allowing for more natural conversations.

sorry not Flash ** Just Gemini 3.5 Live Translate : )

Try it in the @Google Translate app or build in preview via the API in @GoogleAIStudio. Find out more ↓ https://goo.gle/4aEtspk

Gemini 3.5 Live translate: Stream in speech, and stream out the spoken translation.

It also magically works with multiple speakers. It does not work with Klingon (I tried).

Try it on AI Studio: https://aistudio.google.com/live?model=gemini-3.5-live-translate-preview

@OfficialLoganK @GeminiApp @GoogleAIStudio we need Gemini 3.5 flash live speech to text API. Coz the Google Speech to text does not work well for strong accents. Gemini works extremely well with context and Google speech to text only works with local context.

@OfficialLoganK Latency is everything in real-time translation. Sub-200ms feels natural, above that conversations get awkward. 70 languages is impressive. But are all 70 hitting that latency target or is it tiered?

@GoogleDeepMind "cross-language communication" fast is the part that gets me. tried live translate on gemini vs piping audio through gpt and the lag difference was night and day, gemini barely paused between languages.

@OfficialLoganK Bro, when Gemini 3.5 pro? Even beta, it’s already June

@_philschmid No Spanish from Spain, just latin american Spanish

I liked Google because you usually included it, it's a shame you didn't this time because the model does seem to be good

@OfficialLoganK real time speech to speech actually working is the sci fi feature i've been waiting on. the latency on that demo is nuts

@GoogleDeepMind That's a great example, GoogleDeepMind! One caveat is that it still struggles with regional dialects and colloquialisms which can be difficult to model.

@OfficialLoganK and so it starts! frontier labs trying to one-up each others launches.

also this is awesome Logan.

@OfficialLoganK My strategy plan !! 🌐🌟 ⬇️

@OfficialLoganK My next Strategy!!🚀🌐 ⬇️

@GoogleDeepMind Impressive latency for real-time audio translation. This could be a game-changer for global SaaS collaboration—finally breaking language barriers in live meetings.

@GoogleDeepMind wild how "fast, cross-language communication" used to mean a 2-3 second lag that killed any natural flow, now it's actually real-time enough to forget it's there

@OfficialLoganK amazing that we can do this.