Tech • AI • Robotics
Introducing Gemini 3.5 Live Translate
8/10TL;DR
Google has introduced Gemini 3.5 Live Translate, enabling real-time speech-to-speech translation across more than 70 languages via its API.
KEY POINTS
Launch of Gemini 3.5 Live Translate
Google unveiled Gemini 3.5 Live Translate, a new model designed for seamless, real-time voice translation. The system supports over 70 languages for both input and output, allowing developers to integrate multilingual speech capabilities directly into applications.
API Access for Developers
The technology, previously used within Google Translate, is now accessible through the Gemini API and AI Studio. This move expands its use beyond consumer apps, enabling developers to build live translation features into products such as streaming platforms, communication tools, and events.
Live Dubbing Demonstration
One demonstration showcased real-time dubbing of a keynote speech into Hindi, with translated audio streamed directly. The system preserved natural pacing and tone, avoiding delays or robotic interruptions typically associated with live translation tools.
Multilingual Presentation Tool
A second demo highlighted use in live presentations. Speakers can stream their voice, generate a QR code or link, and allow audience members to listen in their preferred language on personal devices. This enables simultaneous multilingual access without additional hardware.
Automatic Language Detection
The model can detect and adapt to language switches automatically, translating speech without manual configuration. Demonstrations included seamless transitions between languages such as German and Sinhala, with consistent output quality.
Natural Speech Output
A key feature is its fluid, natural-sounding audio, with no noticeable pauses or choppiness. The system aims to replicate conversational flow, making translated speech feel continuous and human-like.
Availability Across Platforms
The feature is available in Google Translate on iOS and Android, supports headphone-based use, and is being rolled out in private preview on Google Meet, signaling broader integration into communication tools.
CONCLUSION
Gemini 3.5 Live Translate positions real-time multilingual communication as a built-in capability for apps and services, with potential to reshape global collaboration and accessibility.
Explain this