Google has released Gemini 3.5 Live Translate, a streaming speech-to-speech translation model covering more than 70 languages and 2,000 language combinations. The model generates translated speech continuously within seconds of the speaker, preserving tone, pitch, and pacing — rather than waiting for a complete sentence before translating.
The rollout lands simultaneously across three channels. Consumers can use it now in the Google Translate app on Android and iOS with no sign-up required. Developers get access in public preview through the Gemini Live API and Google AI Studio at $0.023 per minute. Enterprise customers on Google Workspace plans enter a private preview for Google Meet starting this month; Meet's language support jumps from 5 to over 70 with the update, covering more than 2,000 language pair combinations.
On Android, a new listening mode routes translated speech directly through the phone's earpiece, making the feature practical for in-person conversations without putting the phone on speaker. The model detects languages automatically and is designed to work reliably in noisy environments.
