85
AI 摘要
OpenAI在API中正式推出GPT-Realtime-2,这是其迄今为止最智能的语音模型,为语音智能体引入了GPT-5级别的推理能力。该模型使语音智能体能够成为实时协作者,在对话展开过程中聆听、推理并解决复杂问题。同时发布的还有流式模型GPT-Realtime-Translate和GPT-Realtime-Whisper,共同构成了一套面向下一代语音界面的全新音频能力组合。
So impressed by this model…
What will you build with this?
What will we build with this?
Introducing GPT-Realtime-2 in the API: our most intelligent voice model yet, bringing GPT-5-class reasoning to voice agents. Voice agents are now real-time coll...