Google launches Gemini 3.5 Live Translate, a real-time translation model.

Google has officially launched Gemini 3.5 Live Translate, a real-time speech translation model that automatically recognizes over 70 languages and generates fluent translations that preserve the original speaker's intonation, speed, and pitch. The model strikes a balance between waiting for context and instantaneous translation, lags behind the speaker by only a few seconds. The model is available for preview via the Gemini Live API for developers, enterprise users can use it in Google Meet, and general users can experience it through the Google Translate app.