{
  "schemaVersion": "1.0",
  "updates": [],
  "companyNews": [],
  "id": "b780",
  "language": "zh",
  "name": "Hume AI",
  "originalName": "Hume AI",
  "library": "tools",
  "category": "模型开发",
  "categoryKey": "模型开发",
  "categoryKeys": [
    "模型开发"
  ],
  "categories": [
    "模型开发"
  ],
  "entryPoints": [],
  "summary": "评测语音模型与智能体的语调、节奏、打断及上下文表现，从对话生成测试并诊断失败原因。",
  "introduction": "Hume AI 当前的 VoiceEQ 工作流从真实对话识别使用场景，生成测试用例、评价标准和实验计划，再对语音模型或智能体进行压力测试与失败诊断。评测覆盖语调、表达、时序、口音和上下文适配，并结合自动测量与人工判断比较改进。Expression APIs 提供表达分析；这些测量描述信号与交互表现，不直接证明说话人的内在心理状态。",
  "capabilities": [],
  "company": {
    "name": "Hume AI, Inc.",
    "originalName": "Hume AI, Inc.",
    "country": {
      "codes": [
        "US"
      ],
      "names": [
        "美国"
      ],
      "display": "美国",
      "basis": "headquarters",
      "basisLabel": "公司总部"
    }
  },
  "officialUrl": "https://www.hume.ai/",
  "profileUpdatedOn": "2026-10-24",
  "sources": [
    {
      "kind": "product",
      "url": "https://www.hume.ai/"
    }
  ],
  "news": [],
  "links": {
    "interactive": "products/hume-ai--b780.html",
    "reading": "read/zh/products/hume-ai--b780.html",
    "markdown": "read/zh/products/hume-ai--b780.md",
    "json": "read/zh/products/hume-ai--b780.json"
  }
}