Published at 11/3/2026, 2:06:58 AM · 43 views
📰 Hume Launches Audio Decision API: Ask Plain-Language Questions of Raw Audio in One Call, 70.7% Agreement With a Human Panel per Hume
On November 2 voice AI company Hume launched the Audio Decision API, which it calls the first audio-native model that answers questions about speech and sound.
💬 What's your take? Join the discussion in the comments.
Read the original Chinese version
📰 Hume 发布 Audio Decision API：直接对原始音频提问，一次调用判断情绪、说话人数和背景声，官方称与人类评审一致率 70.7%
语音 AI 公司 Hume 11 月 2 日发布 Audio Decision API，号称首个「音频原生」的问答式决策模型：开发者上传原始音频，用一句话写问题并给出候选答案，接口返回每个答案的概率，不经过转写。支持是/否、多选、打分和时间片段四类问题，可判断说话人数、是否有人笑、来电者是否不耐烦、背景是否有警笛等。
💬 你怎么看？欢迎在评论区讨论。
#Hume #Audio Decision API #voice AI
0 Comments