{
  "$schema": "/api/schemas/documents.schema.json",
  "apiVersion": "1.0.0",
  "generatedAt": "2026-07-14T19:46:21.019Z",
  "sourcePath": "src/main/data/docs",
  "docId": "10.5594-MOO-3044",
  "document": {
    "abstract": "A highly manual and time-consuming process has traditionally created descriptive video accessibility tracks, but it is now possible to apply automated software to produce high-quality and nuanced descriptive services. We break down this process including steps for transcription, generative AI scripting using large language models (LLM), generation of synthetic voice, and mixing, and show examples for several different program types. We also demonstrate a heuristic model for descriptive video quality scoring and its application to the AI-generated descriptions created using our methods. Finally, we assess dimensions of quality descriptive video where AI continues to struggle and further technology improvements may be required to match the utility of traditional workflows.",
    "abstract$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "authors": [
      {
        "name": "Bill McLaughlin"
      },
      {
        "name": "James Hu"
      },
      {
        "name": "Fahad Ahmad Arsal"
      },
      {
        "name": "Rhys Fuller"
      }
    ],
    "authors$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "contentType": "orig-research",
    "contentType$meta": {
      "confidence": "high",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "copyright": {
      "holder": "SMPTE",
      "holder$meta": {
        "confidence": "medium",
        "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
        "source": "parsed",
        "updated": "2026-07-10T20:01:10.986Z"
      },
      "year": "2024",
      "year$meta": {
        "confidence": "medium",
        "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
        "source": "parsed",
        "updated": "2026-07-10T20:01:10.986Z"
      }
    },
    "copyright$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "docId": "10.5594-MOO-3044",
    "docId$meta": {
      "confidence": "high",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "docLabel": "MTS 2024, Article 33 (pp. 1 to 13)",
    "docLabel$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "docTitle": "A Fully AI Approach to Descriptive Video Accessibility",
    "docTitle$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "docType": "Conference Paper",
    "docType$meta": {
      "confidence": "high",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "doi": "10.5594/MOO/3044",
    "doi$meta": {
      "confidence": "high",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "href": "https://doi.org/10.5594/MOO/3044",
    "href$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "isbn": {
      "electronic": "978-1-61482-965-2",
      "electronic$meta": {
        "confidence": "medium",
        "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
        "source": "parsed",
        "updated": "2026-07-10T20:01:10.986Z"
      }
    },
    "isbn$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "journalAcronym": "MTS",
    "journalAcronym$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "journalTitle": "SMPTE 2024 Media Technology Summit",
    "journalTitle$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "keywords": [
      "Descriptive Video",
      "Audio Description",
      "Accessibility",
      "Generative AI",
      "Large Language Models (LLMs)"
    ],
    "keywords$meta": {
      "confidence": "high",
      "note": "Long-tail keyword cleanup (keywordLongTailApply.js): prose/contentless terms dropped, acronym casing + typos fixed, duplicate variants folded; all surviving terms indexed in controlledKeywords.",
      "source": "resolved",
      "updated": "2026-07-14T16:29:38.680Z"
    },
    "pages": "1–13",
    "pages$meta": {
      "confidence": "high",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "publicationDate": "2024-10-21",
    "publicationDate$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "publisher": "SMPTE",
    "publisher$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "publisherLocation": {
      "city": "Hollywood, CA",
      "city$meta": {
        "confidence": "medium",
        "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
        "source": "parsed",
        "updated": "2026-07-10T20:01:10.986Z"
      }
    },
    "publisherLocation$meta": {
      "confidence": "medium",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    },
    "status": {
      "active": true,
      "active$meta": {
        "confidence": "medium",
        "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
        "source": "parsed",
        "updated": "2026-07-10T20:01:10.986Z"
      }
    },
    "volume": "00",
    "volume$meta": {
      "confidence": "high",
      "note": "Ingested from SMPTE canonical content_batch via ingestNlmCanonicalDocs.js",
      "source": "parsed",
      "updated": "2026-07-10T20:01:10.986Z"
    }
  }
}