docling/tests/data/groundtruth/docling_v2/webvtt_example_02.vtt.json

{
  "schema_name": "DoclingDocument",
  "version": "1.8.0",
  "name": "webvtt_example_02",
  "origin": {
    "mimetype": "text/vtt",
    "binary_hash": 5029965721282070624,
    "filename": "webvtt_example_02.vtt"
  },
  "furniture": {
    "self_ref": "#/furniture",
    "children": [],
    "content_layer": "furniture",
    "name": "_root_",
    "label": "unspecified"
  },
  "body": {
    "self_ref": "#/body",
    "children": [
      {
        "$ref": "#/groups/0"
      },
      {
        "$ref": "#/groups/2"
      },
      {
        "$ref": "#/groups/4"
      },
      {
        "$ref": "#/groups/6"
      }
    ],
    "content_layer": "body",
    "name": "_root_",
    "label": "unspecified"
  },
  "groups": [
    {
      "self_ref": "#/groups/0",
      "parent": {
        "$ref": "#/body"
      },
      "children": [
        {
          "$ref": "#/texts/0"
        },
        {
          "$ref": "#/groups/1"
        }
      ],
      "content_layer": "body",
      "name": "WebVTT cue block",
      "label": "section"
    },
    {
      "self_ref": "#/groups/1",
      "parent": {
        "$ref": "#/groups/0"
      },
      "children": [
        {
          "$ref": "#/texts/1"
        },
        {
          "$ref": "#/texts/2"
        }
      ],
      "content_layer": "body",
      "name": "WebVTT cue voice span",
      "label": "inline"
    },
    {
      "self_ref": "#/groups/2",
      "parent": {
        "$ref": "#/body"
      },
      "children": [
        {
          "$ref": "#/texts/3"
        },
        {
          "$ref": "#/groups/3"
        }
      ],
      "content_layer": "body",
      "name": "WebVTT cue block",
      "label": "section"
    },
    {
      "self_ref": "#/groups/3",
      "parent": {
        "$ref": "#/groups/2"
      },
      "children": [
        {
          "$ref": "#/texts/4"
        },
        {
          "$ref": "#/texts/5"
        }
      ],
      "content_layer": "body",
      "name": "WebVTT cue voice span",
      "label": "inline"
    },
    {
      "self_ref": "#/groups/4",
      "parent": {
        "$ref": "#/body"
      },
      "children": [
        {
          "$ref": "#/texts/6"
        },
        {
          "$ref": "#/groups/5"
        },
        {
          "$ref": "#/texts/9"
        }
      ],
      "content_layer": "body",
      "name": "WebVTT cue block",
      "label": "section"
    },
    {
      "self_ref": "#/groups/5",
      "parent": {
        "$ref": "#/groups/4"
      },
      "children": [
        {
          "$ref": "#/texts/7"
        },
        {
          "$ref": "#/texts/8"
        }
      ],
      "content_layer": "body",
      "name": "WebVTT cue voice span",
      "label": "inline"
    },
    {
      "self_ref": "#/groups/6",
      "parent": {
        "$ref": "#/body"
      },
      "children": [
        {
          "$ref": "#/texts/10"
        },
        {
          "$ref": "#/groups/7"
        }
      ],
      "content_layer": "body",
      "name": "WebVTT cue block",
      "label": "section"
    },
    {
      "self_ref": "#/groups/7",
      "parent": {
        "$ref": "#/groups/6"
      },
      "children": [
        {
          "$ref": "#/texts/11"
        },
        {
          "$ref": "#/texts/12"
        }
      ],
      "content_layer": "body",
      "name": "WebVTT cue voice span",
      "label": "inline"
    }
  ],
  "texts": [
    {
      "self_ref": "#/texts/0",
      "parent": {
        "$ref": "#/groups/0"
      },
      "children": [],
      "content_layer": "body",
      "label": "text",
      "prov": [],
      "orig": "00:00.000 --> 00:02.000",
      "text": "00:00.000 --> 00:02.000"
    },
    {
      "self_ref": "#/texts/1",
      "parent": {
        "$ref": "#/groups/1"
      },
      "children": [],
      "content_layer": "body",
      "label": "text",
      "prov": [],
      "orig": "Esme (first, loud): ",
      "text": "Esme (first, loud): "
    },
    {
      "self_ref": "#/texts/2",
      "parent": {
        "$ref": "#/groups/1"
      },
      "children": [],
      "content_layer": "body",
      "label": "text",
      "prov": [],
      "orig": "It’s a blue apple tree!",
      "text": "It’s a blue apple tree!",
      "formatting": {
        "bold": false,
        "italic": false,
        "underline": false,
        "strikethrough": false,
        "script": "baseline"
      }
    },
    {
      "self_ref": "#/texts/3",
      "parent": {
        "$ref": "#/groups/2"
      },
      "children": [],
      "content_layer": "body",
      "label": "text",
      "prov": [],
      "orig": "00:02.000 --> 00:04.000",
      "text": "00:02.000 --> 00:04.000"
    },
    {
      "self_ref": "#/texts/4",
      "parent": {
        "$ref": "#/groups/3"
      },
      "children": [],
      "content_layer": "body",
      "label": "text",
      "prov": [],
      "orig": "Mary: ",
      "text": "Mary: "
    },
    {
      "self_ref": "#/texts/5",
      "parent": {
        "$ref": "#/groups/3"
      },
      "children": [],
      "content_layer": "body",
      "label": "text",
      "prov": [],
      "orig": "No way!",
      "text": "No way!",
      "formatting": {
        "bold": false,
        "italic": false,
        "underline": false,
        "strikethrough": false,
        "script": "baseline"
      }
    },
    {
      "self_ref": "#/texts/6",
      "parent": {
        "$ref": "#/groups/4"
      },
      "children": [],
      "content_layer": "body",
      "label": "text",
      "prov": [],
      "orig": "00:04.000 --> 00:06.000",
      "text": "00:04.000 --> 00:06.000"
    },
    {
      "self_ref": "#/texts/7",
      "parent": {
        "$ref": "#/groups/5"
      },
      "children": [],
      "content_layer": "body",
      "label": "text",
      "prov": [],
      "orig": "Esme: ",
      "text": "Esme: "
    },
    {
      "self_ref": "#/texts/8",
      "parent": {
        "$ref": "#/groups/5"
      },
      "children": [],
      "content_layer": "body",
      "label": "text",
      "prov": [],
      "orig": "Hee!",
      "text": "Hee!",
      "formatting": {
        "bold": false,
        "italic": false,
        "underline": false,
        "strikethrough": false,
        "script": "baseline"
      }
    },
    {
      "self_ref": "#/texts/9",
      "parent": {
        "$ref": "#/groups/4"
      },
      "children": [],
      "content_layer": "body",
      "label": "text",
      "prov": [],
      "orig": "laughter",
      "text": "laughter",
      "formatting": {
        "bold": false,
        "italic": true,
        "underline": false,
        "strikethrough": false,
        "script": "baseline"
      }
    },
    {
      "self_ref": "#/texts/10",
      "parent": {
        "$ref": "#/groups/6"
      },
      "children": [],
      "content_layer": "body",
      "label": "text",
      "prov": [],
      "orig": "00:06.000 --> 00:08.000",
      "text": "00:06.000 --> 00:08.000"
    },
    {
      "self_ref": "#/texts/11",
      "parent": {
        "$ref": "#/groups/7"
      },
      "children": [],
      "content_layer": "body",
      "label": "text",
      "prov": [],
      "orig": "Mary (loud): ",
      "text": "Mary (loud): "
    },
    {
      "self_ref": "#/texts/12",
      "parent": {
        "$ref": "#/groups/7"
      },
      "children": [],
      "content_layer": "body",
      "label": "text",
      "prov": [],
      "orig": "That’s awesome!",
      "text": "That’s awesome!",
      "formatting": {
        "bold": false,
        "italic": false,
        "underline": false,
        "strikethrough": false,
        "script": "baseline"
      }
    }
  ],
  "pictures": [],
  "tables": [],
  "key_value_items": [],
  "form_items": [],
  "pages": {}
}