AI Read the original on ITC.ua 2 min read 0

Google launches Gemini 3.5 Live Translate for real-time communication

Google has unveiled Gemini 3.5 Live Translate, a sophisticated new model designed to provide seamless, real-time speech translation globally. This advanced system supports over 70 languages and enables more than 2,000 language combinations in single sessions. Unlike older systems that pause between phrases, this technology generates continuous audio streams, drastically improving conversational flow for international users.

Екран мобільного пристрою відображає яскравий інтерфейс з логотипом Google і назвою штучного інтелекту Gemini.
Екран мобільного пристрою відображає яскравий інтерфейс з логотипом Google і назвою штучного інтелекту Gemini. · Image source: ITC.ua

Google has officially announced Gemini 3.5 Live Translate, positioning it as a major leap forward in real-time speech translation capabilities. The model is designed to deliver smooth and natural-sounding translated speech while accurately preserving the speaker's original intonation, tempo, and pitch.

According to ITC.ua, this new generation of AI overcomes the limitations of traditional sequential translation systems. Instead of waiting for a speaker to finish an entire sentence before responding—a process that often introduces awkward pauses—Gemini 3.5 continuously generates audio output. This continuous flow allows users to stay synchronized with the original speaker throughout the conversation.

Advancements in Real-Time Conversational Flow

The core innovation of Gemini 3.5 lies in its ability to balance two competing demands: maintaining high translation quality by waiting for sufficient context, and ensuring immediate synchronization with the speaker. Google notes that the model achieves this delicate balance through continuous generation.

This powerful functionality is already rolling out across several platforms:

  • Mobile Integration: The feature is available in Google Translate for both Android and iOS. On Android, a new "listening mode" allows users to hear translations via the phone's speaker during calls without needing headphones.
  • Google Meet Expansion: Previously limited to only five languages, Gemini 3.5 Live Translate dramatically expands support within video conferencing applications, enabling over 2,000 language combinations in one meeting.
  • Web Access: A new dedicated button has been added to the web interface to instantly initiate speech translation sessions.

Broadening Accessibility and Enterprise Use

The rollout strategy for Gemini 3.5 is phased, beginning with a closed preview for select Google Workspace business clients this month. The wider public launch is scheduled for later in the year. Beyond consumer applications, the technology is being made available to developers through the Gemini Live API and Google AI Studio.

This developer access allows third-party creators to integrate advanced multilingual capabilities into various services

FAQ

How does Gemini 3.5 improve real-time conversations?
Unlike older systems that pause between phrases, Gemini 3.5 continuously generates audio output. This continuous flow allows users to stay synchronized with the original speaker throughout the conversation while maintaining high translation quality.
Where can I access the Gemini 3.5 Live Translate feature?
The technology is available in Google Translate for both Android and iOS, and it expands support within Google Meet video conferencing applications. A new dedicated button also exists on the web interface.
When will the general public be able to use this translation model?
The rollout strategy begins with a closed preview for select Google Workspace business clients this month. The wider public launch of Gemini 3.5 Live Translate is scheduled to occur later in the year.
Telegram

Fresh news on our Telegram

Get instant alerts for new posts in «AI»

@proaiandevenmore