Google introduces Gemini 3.5 Live Translate technology

Google has announced a new Gemini 3.5 Live Translate audio model designed for real-time speech translation. This technology aims to make communication between people speaking different languages more natural and seamless. Unlike traditional systems that work in stages, the new model performs continuous streaming translation. This is reported by Ixbt.com reports .

Developers note that Gemini 3.5 Live Translate generates translations while the speaker is still talking, with a latency of just a few seconds. This approach helps avoid awkward pauses in conversation and maintains the rhythm of the dialogue. The system can automatically detect over 70 languages and preserves the speaker's intonation, pace, and volume.

The technology is being gradually integrated into Google products. For corporate clients, the update will appear in Google Meet. For the general user base, this feature will be available in the Google Translate app on Android and iOS. The model also stands out for its resistance to background noise.

A special "listening mode" is also being introduced for Android device owners. In this mode, users can hold the phone to their ear like a regular call and listen to the translated audio stream directly through the speaker. When using headphones, users will be able to hear even more fluent and natural speech.