The Information Machine

Fascinating. Google just released Gemini 3.5 Live Translate.

Rohan Paul Twitter · Rohan Paul (@rohanpaul_ai) · 2026-06-09

Google released Gemini 3.5 Live Translate, a real-time speech-to-speech translation model that begins outputting the translated language while the source speaker is still talking, surpassing traditional sentence-level translation delays.

Open original ↗

Appears in

Extraction

Topics: google-geminireal-time-translationspeech-aimultimodal-ai

Claims

  • Google released Gemini 3.5 Live Translate as a real-time speech-to-speech translation model.
  • The model begins speaking the translation before the original speaker has finished their sentence.
  • Traditional translation systems wait for a complete sentence before beginning output because early words can change meaning.

Key quotes

A live speech-to-speech translation model that starts speaking in another language while the original speaker is still talking.
Older translation systems often wait for a full sentence, because early words can be [ambiguous].