Google releases Gemini 3.5 Live Translate, a real-time audio translation model for 70+ languages. The model detects languages automatically and, according to Google, preserves the speaker's tone, pace, and pitch. It also translates continuously without waiting for a sentence to end.
Gemini 3.5 Live Translate is available now for developers through the Gemini Live API and Google AI Studio, as a preview for businesses in Google Meet, and for all users in the Google Translate app on Android and iOS. In Google Meet, language support jumps from five to over 70 languages with more than 2,000 language combinations. Ride-hailing service Grab is reportedly testing the model for driver-passenger communication. All generated audio is tagged with an inaudible SynthID watermark.
AI News Without the Hype – Curated by Humans
Subscribe to THE DECODER for ad-free reading, a weekly AI newsletter, our exclusive "AI Radar" frontier report six times a year, full archive access, and access to our comment section.