Synthszr Charts — die großen AI-Marken im Wettkampf ums Podium
synthszr charts
google

Gemini 3.5 Live Translate

#1 in AI Translation

google · v3.5 · live translate · seit Juni 2026 · 16× · zuletzt 02. Juli 2026

100
Momentum

Gemini 3.5 Live Translate is Google's audio model for real-time speech translation. The model automatically detects 70+ languages and generates translated speech that preserves the speaker's intonation, pace, and pitch. Unlike turn-by-turn systems, it produces speech continuously with only a few seconds of lag while the speaker is still talking.

Momentum trend
04.04.03.07.

Features

LatencyNear real-time; according to Google, the model stays "mere seconds" behind the speaker (continuous streaming rather than turn-by-turn). Input audio is streamed in 100 ms chunks.
LicenseProprietary (Google). API usage is subject to the Gemini API Additional Terms of Service. No open-source release.
PriceGoogle Translate app (consumer): free. Gemini Live API (developer): approx. $0.023 per minute of audio (token-based: 25 tokens/second of audio). Google Meet (enterprise): pricing not yet announced.
Release DateJune 9, 2026
Languages70+ languages, automatically detected (including fast language switching between speakers); over 2,000 language combinations possible in a single session.

Sources (16)

Subscribe free. Unsubscribe the second it sucks.

High-signal news across AI, business, UX, and tech. Every morning.