elvis· @omarsar0 · X·· 3 小时前AI 评分58
AI 导读
Hume 发布 Audio Decision API，可用一次调用回答说话人数、是否有人发笑、来电者是否沮丧等问题，官方称过去需要 10+ 个模型。该 API 基于 1000 万+ 人类判断训练，支持实时与数小时录音分析，目前需申请访问权限。elvis 认为其音频原生特性为语音智能体交互带来新可能。
正文 · 原文
One of the most insane applications of decision models I've seen so far.
Unlike other decision models, this one is audio-native.
This unlocks new ways for how we interact with agents.
The Hume Audio Decision API from @hume_ai listens to the audio itself. An agent can ask questions like "Is the caller frustrated?" in one call and respond to how you said something.
Tons of obvious business applications here. But I think this has huge applications in personal agents that we are increasingly interacting with via voice.
Introducing the Hume Audio Decision API. Ask anything: How many speakers? Is anyone laughing? Is the caller frustrated? What used to take 10+ models is now one call. Trained on 10M+ human judgments, so it answers the way human listeners would. Realtime, or across hours of recordings. Request access below.在 X 查看被引用的帖子
来源：elvis · x.com