Google on Tuesday, June 9, 2026, unveiled Gemini 3.5 Live Translate, a new audio model designed to translate conversations in real time, aiming to make cross-language communication more natural. The tool is part of the Gemini 3.5 update, which focuses on improving audio processing and latency.
According to Google's announcement, the model can process speech and generate translations with minimal delay, allowing users to speak and hear translations almost simultaneously. This is a significant step beyond traditional text-based translation tools, which often require pauses between speaking and translation.
Gemini 3.5 Live Translate is expected to support multiple languages, though the initial rollout details and specific language support have not been fully disclosed. The feature is integrated into the Gemini app and is available for select users starting this week.
Early reviews from tech journalists who tested the feature highlight its ability to handle natural speech patterns, including interruptions and overlapping dialogue, which are common in real conversations. However, accuracy may vary depending on language pairs and background noise.
