Google Meet translates in real-time while preserving the speaker’s voice

Google has introduced a revolutionary real-time speech translation feature for its Meet video conferencing service. I am pleasantly shocked! The main feature of this technology is the ability to not only translate what is said, but also preserve the individual characteristics of the speaker’s voice. Including timbre, intonations, and emotional nuances.

The system works based on Gemini artificial intelligence and allows conversation participants to communicate across the language barrier, while maintaining the naturalness of communication. At the I/O conference, Google demonstrated this technology in action. In the demonstration, an English-speaking participant communicated with a colleague in Spanish. After activating the Gemini function, the system instantly began translating speech in both directions. And each participant heard the interlocutor in their native language while preserving the original voice characteristics.

Currently, the technology works only with English and Spanish languages, but the company has already announced plans to add support for Italian, German, and Portuguese languages in the coming weeks. Now the function is available to paid Google Meet subscribers, and by the end of the year, they plan to open it to corporate clients.

This technology potentially can change the approach to international communications, making them more natural and effective for business, education, and personal communication with foreign friends or relatives.

I am really surprised! Google Meet simultaneous translation focuses not only on the accuracy of meaning transmission, but also on preserving the paralinguistic elements of speech. Which makes communication between speakers of different languages as close as possible to natural communication. How cool it will be to talk to any person on the planet in any language. I’m really looking forward to this technology being widely available.