Google Launches Gemini 3.5 Live Translate for Instant Voice Translation

Google has released Gemini 3.5 Live Translate, a new artificial intelligence tool designed for instant voice-to-voice translation. This innovation promises broader accessibility and reduced latency across more than 70 languages, the company announced.

This development builds on Google's long-standing pursuit of real-time translation, which it describes as one of its "pioneering machine learning experiments." Historically, demonstrations of this technology at Google events often required specific hardware, such as Google phones or earbuds.

Last year, Google already made real-time translation available to more users through its Translate app. Now, with Gemini 3.5 Live Translate, this capability is expanding further, allowing individuals to access instant translation in a wider range of settings.

The core of this advancement is a new AI model, part of the Gemini 3.5 family unveiled at the I/O event. While only the Flash version of this series had been released until now, a more powerful Pro model is anticipated "in the coming weeks."

Gemini 3.5 Live Translate is a speech-to-speech model tuned to automatically detect and translate in more than 70 languages. This significant linguistic coverage aims to break down communication barriers more effectively.

Google states that the model operates quickly enough to keep pace with a normal conversation, following the speaker by just a few seconds. Crucially, it also focuses on matching the intonation, pacing, and pitch of the original voice.

This emphasis on natural voice replication means the translated output sounds more like the speaker and less like a generic robotic voice. While the showcased demos are recorded under controlled conditions, their performance appears impressive.

For users, this implies a substantial improvement in the fluidity of multilingual interactions. The reduced reliance on specific devices makes this advanced technology more accessible to a broader audience, fostering seamless communication.

We won't have to wait long to personally verify the model's capabilities. The promise of instant, natural translation across numerous languages opens up vast possibilities for global communication.