Gemini 3.5 Live Translate rolls out near real-time speech translation
*Google adds fluid voice translation to AI Studio, Translate, and Meet through its latest Gemini model.*
Google announced Gemini 3.5 Live Translate, a feature that delivers near real-time, natural speech translation. The capability appears in Google AI Studio, Google Translate, and Google Meet.
The update focuses on spoken language rather than text. It processes audio input and produces spoken output with reduced lag and more conversational phrasing than earlier translation systems.
Two announcements from Google and Google DeepMind describe the same rollout. Both list the three product surfaces and emphasize the shift toward spoken, low-latency results.
No performance benchmarks or side-by-side comparisons appear in the releases. The announcements contain no on-record quotes from engineers or product leads.
Why it matters
Engineers and teams already inside Google’s ecosystem gain an incremental improvement in cross-language calls and meetings. The change matters most to users who need spoken rather than typed translation and who stay within the three named Google products. Outside those surfaces the feature does not yet exist.
---
Sources:
No comments yet