{
  "@context": "https://doi.org/10.5063/schema/codemeta-2.0",
  "@type": "SoftwareSourceCode",
  "name": "HEAR: Vision-Sound-Language-Action for Sound-Centric Manipulation",
  "description": "Reference implementation of HEAR, a Vision-Sound-Language-Action framework for sound-centric robot manipulation, together with the HEAR-Bench benchmark.",
  "codeRepository": "https://github.com/IRMVLab/HEAR",
  "url": "https://hear.irmv.top/",
  "license": "https://www.apache.org/licenses/LICENSE-2.0",
  "programmingLanguage": [
    "Python"
  ],
  "datePublished": "2026-03-17",
  "dateModified": "2026-09-16",
  "author": [
    {
      "@type": "Person",
      "givenName": "Nie",
      "familyName": "Chang",
      "affiliation": {
        "@type": "Organization",
        "name": "Shanghai Jiao Tong University"
      }
    },
    {
      "@type": "Person",
      "givenName": "Deng",
      "familyName": "Tianchen",
      "affiliation": {
        "@type": "Organization",
        "name": "Shanghai Jiao Tong University"
      }
    },
    {
      "@type": "Person",
      "givenName": "Wang",
      "familyName": "Guangming",
      "affiliation": {
        "@type": "Organization",
        "name": "University of Cambridge"
      }
    },
    {
      "@type": "Person",
      "givenName": "Liu",
      "familyName": "Zhe",
      "affiliation": {
        "@type": "Organization",
        "name": "Shanghai Jiao Tong University"
      }
    },
    {
      "@type": "Person",
      "givenName": "Wang",
      "familyName": "Hesheng",
      "affiliation": {
        "@type": "Organization",
        "name": "Shanghai Jiao Tong University"
      }
    }
  ],
  "referencePublication": {
    "@type": "ScholarlyArticle",
    "name": "Towards the Vision-Sound-Language-Action Paradigm: The HEAR Framework for Sound-Centric Manipulation",
    "url": "https://arxiv.org/abs/2603.16086",
    "sameAs": "https://arxiv.org/abs/2603.16086"
  },
  "keywords": [
    "robot manipulation",
    "Vision-Sound-Language-Action",
    "VSLA",
    "sound-centric manipulation",
    "robot audition",
    "audio world model",
    "action chunking",
    "embodied AI",
    "physical AI"
  ],
  "developmentStatus": "active"
}