ReadenReaden

Google Unveils Gemini 3.5 Live Translate: Real-Time Speech Translation for 70+ Languages, Available in Google Translate App

科技文章初级 · 3.0
610 词 4 分钟 15 次阅读
#AI #科技

Gemini 3.5 Live Translate can automatically recognize over 70 languages and generate fluent, natural-sounding translated speech while preserving the speaker's tone, pace, and pitch.

On June 9 (Beijing time), Google officially unveiled its latest real-time speech-to-speech translation audio model, Gemini 3.5 Live Translate, marking the next step forward in real-time translation.

Gemini 3.5 Live Translate can automatically recognize over 70 languages and generate fluent, natural-sounding translated speech while preserving the speaker's tone, pace, and pitch.

According to Google, traditional turn-based systems typically wait for the speaker to finish before beginning translation. Gemini 3.5 Live Translate, however, generates speech continuously, striking a balance between "waiting for more context to improve translation quality" and "translating instantly to keep pace with the speaker." Throughout the conversation, the translated audio remains smooth, avoiding awkward pauses, and consistently lags only a few seconds behind the speaker.

Gemini 3.5 Live Translate is rolling out across Google products starting today:

Developers can access a public preview through the Gemini Live API and Google AI Studio.

Enterprise customers can use a private preview in Google Meet starting this month.

Regular users can access it through the Google Translate app on Android and iOS.

Gemini 3.5 Live Translate processes content synchronously during speech streaming, making cross-language communication smoother. The model can automatically handle multilingual input without requiring manual setting adjustments, and its noise-resistance capabilities help applications adapt to noisy, complex, and unpredictable environments. Developers can use Gemini 3.5 Live Translate for multilingual calls, meetings, classes, live streams, and broadcasts, delivering a real-time interpretation experience.

Gemini 3.5 Live Translate will be available on the Google Translate app for Android and iOS worldwide. When using the real-time translation feature, users simply need to connect any headphones to enjoy smoother translations across over 70 languages, with the translated speech more closely matching the speaker's tone.

Android users will also gradually receive a new "Listening Mode" powered by 3.5 Live Translate. Users can hold the phone to their ear like taking a phone call and listen to the translated audio directly through the earpiece speaker. This mode is more convenient when you don't have headphones and don't want others around you to hear the translation.

评论

0 条讨论

按时间

登录后发表评论

立即登录

暂无评论

成为第一个分享想法的人吧!