Google announced real-time speech translation is coming to Google Meet. It uses a DeepMind large-language audio model to let you have natural, free-flowing conversations with people who speak different languages.

Speech translation in Meet translates spoken words into the listener’s preferred language in real time. Voice, tone, and expression are all preserved in the translation.

Google says this feature can help global teams stay connected by letting colleagues in different regions talk and understand each other in real time. The translation works with very low delay, so several people can join the same conversation naturally something that wasn’t possible before.

When someone speaks in Google Meet, you’ll still hear their original voice softly in the background while the translated speech plays on top. This layered audio preserves natural intonation and emotion, making conversations feel authentic. You can adjust the mix between the original and translated audio in your settings, and live captions appear below the video to help everyone follow along seamlessly.

The feature is rolling out first in a consumer beta, supporting English and Spanish, with additional languages like Italian, German, and Portuguese arriving soon. Business users on Google Workspace will gain early access as well, allowing global teams to hold secure, real-time multilingual meetings without relying on third-party tools or manual interpretation.

In conclusion, Google’s real-time speech translation in Meet represents a major step forward in making virtual conversations seamless and inclusive across language barriers. By preserving voice, tone, and expression while delivering near-instant translations, the feature ensures natural, engaging dialogue for personal and professional use. As support expands to more languages and rolls out to Workspace customers, businesses and individuals worldwide will be able to connect effortlessly, fostering stronger relationships and more effective collaboration in an increasingly globalized world.