With the rise of artificial intelligence in everyday tools, Google Meet has taken a major leap forward by eliminating one of the most persistent challenges in global communication: the language barrier. Thanks to Gemini, Google’s advanced AI model, the platform now offers real-time dubbing during video calls, enabling participants to speak and hear in their native languages while maintaining the authenticity of the original voice.
A New Era of Multilingual Communication
Previously, Google Meet offered features like live captions in multiple languages and AI-powered note-taking. However, the recent introduction of real-time voice translation and dubbing represents a qualitative leap. This innovation allows up to 100 participants to engage in virtual meetings where their spoken words are translated almost instantly into another language — with tone, rhythm, and vocal expression preserved.
As Javier Martínez, Director of Google Cloud’s Pre-Sales team, explained in an interview with Cadena SER: “Until now, the system learned your voice and could synthesize it to sound similar. But with this model, we translate your audio and return it directly. The revolution is that we now translate audio-to-audio.”
How the Dubbing Feature Works
The feature is intuitive and easy to activate. From the usual Google Meet interface, users can simply enable the real-time translation function — much like turning on subtitles. Once enabled, the conversation becomes bidirectional: for example, if one participant speaks in English, the system provides a spoken translation in Spanish and vice versa, with only a slight delay of a few seconds.
The original voice remains in the background, while the dubbed version takes the foreground. This allows the translated voice to carry the same tone, rhythm, and emotional nuances, offering a far more natural and immersive experience than traditional translation tools.
Powered by Gemini: AI That Understands and Feels
At the core of this breakthrough lies Gemini — Google’s cutting-edge AI designed to process and synthesize language. Unlike conventional translators that rely on first transcribing and then translating text, Gemini interprets and translates directly from audio to audio. This unique process enables real-time dubbing that is not only fast but emotionally intelligent.
During a demonstration at Google’s Shoreline Amphitheater in California, the company showcased a video where an American and a Latina woman discussed renting a house — each speaking in their native language. With Gemini, the conversation flowed naturally, with neither party needing to compromise on language.
A Seamless User Experience
Currently available in beta for subscribers of Google Meet’s AI Pro and Ultra plans, this feature supports Spanish and English. Google has confirmed that more languages will be added in the coming weeks.
Importantly, users don’t need to download extra software or make complex configurations. The translation function is built directly into the Meet interface, making it accessible with just a few clicks.
This simplicity, combined with the sophisticated AI voice synthesis, sets Google Meet apart from traditional language tools. Unlike robotic or impersonal voiceovers, Gemini mimics the vocal and emotional traits of each speaker, ensuring that the message is delivered with authenticity.
Toward a More Connected World
This real-time dubbing function doesn’t interrupt the flow of conversation or distort the speaker’s original style — even in spontaneous dialogues or sensitive negotiations. Google’s internal tests have shown that the system successfully maintains the personal expression of each speaker, fostering trust and clarity in cross-linguistic communication.
With plans to roll out this feature to business users and expand the language options, Google is laying the groundwork for a more inclusive and global virtual meeting space. By bringing people closer across linguistic boundaries, Google Meet is not just facilitating communication — it’s redefining what it means to truly understand one another.