Gemini 3.5 Live Translate Brings Seamless Real-Time Speech Translation

Google continues to improve its AI-powered language tools. Now, the company has introduced Gemini 3.5 Live Translate, a new audio model that delivers near real-time speech-to-speech translation in more than 70 languages. The model is currently rolling out across several products, including the Google Translate app.

Google’s latest audio model delivers continuous live voice translation

Based on Gemini 3 Pro, Gemini 3.5 Live Translate generates smooth, natural-sounding translated speech across languages. Unlike traditional translation tools that wait for a speaker to finish talking, this new model performs in real time. It balances two trade-offs. One is waiting for enough context to improve translation quality. The other is translating immediacy to stay in sync with the speaker.

As a result, Gemini 3.5 Live Translate offers seamless audio without awkward pauses and lasts only a few seconds behind the speaker during the conversation. The Google Translate app on Android and iOS now supports this model. In Live Translate mode, you can connect any pair of headphones to get a more natural speech-to-speech translation experience.

On Android, Google also brings a new “listening mode” with Gemini 3.5 Live Translate. This feature plays translated speech through the phone’s earpiece. You just need to hold the phone to their ear and hear the translation immediately. This comes in handy when you need quick, private translations.

Furthermore, Google Meet will soon get support for Gemini 3.5 Live Translate, offering real-time speech translation in your video meetings. The company will bring the new experience in a private preview for select business Google Workspace customers starting this month. It promises to make the new model available widely later this year.

Google adds a SynthID watermark to all audio generated by the model. It helps detect AI-generated content and prevents misinformation. Meanwhile, Google is rolling out Gemini 3.5 Live Translate to its other products for developers. It is available in public preview through the Gemini Live API and Google AI Studio.