Google AI released Gemini 3.5 Live Translate, supporting 70+ languages with near real-time latency speech-to-speech translation. This model directly proces...
Google AI released Gemini 3.5 Live Translate, supporting 70+ languages with near real-time latency
speech-to-speech translation. This model directly processes the raw audio stream, preserving the speaker's intonation, rhythm, and pitch. Southeast Asia's super app Grab is exploring cross-language communication between drivers and passengers, with users initiating over 10 million voice calls each month. Developers can build applications by integrating LiveKit, Fishjam, Pipecat, or Vision Agents via the Gemini Live API. LiveKit has enabled multilingual real-time understanding in virtual meeting rooms; Software Mansion breaks through streaming bottlenecks through the MoQ protocol; VisionAgents AI demonstrated dynamic multilingual switching capabilities. Developers can try out and access Cookbook sample code in Google AI Studio.
Google AI 发布 Gemini 3.5 Live Translate，支持 70+ 语言、近实时延迟的语音到语音翻译。
该模型直接处理原始音频流，保留说话者语调、节奏和音高。
东南亚超级应用 Grab 正探索将其用于司机与乘客间的跨语言沟通，其用户每月发起超 1000 万次语音通话。
开发者可通过 Gemini Live API 集成 LiveKit、Fishjam、Pipecat 或 Vision Agents 构建应用。
LiveKit 已实现虚拟会议室多语言即时理解；
Software Mansion 结合 MoQ 协议突破流媒体瓶颈；
VisionAgents AI 展示了动态多语言切换能力。
开发者可在 Google AI Studio 试用并获取 Cookbook 示例代码。
Google AI 发布 Gemini 3.5 Live Translate，可在 70 多种语言间进行近实时语音到语音翻译，并直接处理原始音频流，以保留说话者的语调、节奏和音高。
Grab 正探索将该模型用于司机与乘客的跨语言沟通，其用户每月发起超过 1000 万次语音通话。开发者可通过 Gemini Live API 结合多种框架构建应用。
Aioga 判断，此次更新的重点不仅是覆盖语言数量，还包括语音特征保留和流式集成能力。其实际价值仍需结合真实通话中的延迟、稳定性和切换表现观察。
该能力可能降低会议、出行沟通和流媒体场景的语言障碍。LiveKit、Software Mansion 与 VisionAgents AI 的案例值得关注，但材料未提供规模化部署效果。 建议开发者先在 Google AI Studio 试用并参考 Cookbook 示例代码，再围绕目标语言、实时延迟、语音特征保留及动态语言切换开展场景验证。
The readable text on this page was extracted from the public source and organized with attribution, publication time and the original link. Copyright remains with the original author and publisher.
Ingestion channel: Summary aggregation · Source domain: x.com
Source: X：Google AI for Developers (@googleaidevs)
Original link: Open original source
Aioga archive: Open intelligence page
Content record: social-summary · Updated: 2026-07-14T15:17:40.000Z
统一接入主流 AI 模型 API，为开发、测试与生产环境提供稳定调用入口。
立即访问 api.w173.com
Scan to open this articleAioga aggregates global AI updates and preserves source information for verification and citation.