By Maria Stasimioti
Publication Date: 2025-11-21 09:30:00
On November 19, 2025, Google revealed the technical details of the real-time speech-to-speech translation (S2ST) system that already powers Google Meet and the Pixel 10.
The company outlined an end-to-end streaming architecture that can translate speech into the original speaker’s voice with a delay of about two seconds – significantly faster than cascaded pipelines.
The research paper does not announce a new product. Instead, Google is revealing the architecture that has been quietly running in Google Meet and Pixel 10 devices for months.
During Google’s annual developer conference, Google I/O 2025, held May 20-21, 2025, the company announced real-time English to Spanish speech-to-speech translation in Google Meet.
The company did not disclose any technical details at this time. The demo video shown to users in May is the same one embedded in the new post – a strong indication that the underlying system has been around for months. The feature was initially introduced…