Google recently unveiled Gemini 3.5 Live Translate, a new audio model designed to make real-time multilingual conversations feel more natural. Unlike traditional translation systems that process speech in turns, Gemini 3.5 Live Translate continuously listens, translates and speaks, allowing conversations to flow with only a few seconds of delay. The model automatically detects spoken languages and supports more than 70, enabling thousands of language pairs within a single conversation. Google says the technology is now available to developers and partners, who can integrate it into meetings, communication platforms and mobile applications. The biggest change is continuous streaming translation, resulting in a more fluid conversational experience with fewer awkward pauses, interruptions and delays. Google said it can operate in noisy environments and is designed to handle background sounds, overlapping voices and informal speech patterns. The system attempts to preserve elements of the original speaker's delivery, including pacing, intonation and emotional tone.