Bookolot deep dive into books

страница 26

26 / 44 русская дореформенная уверенность 0.70 человеком не проверено

Провенанс

transcript transcript
2026-09-29T22:02:25+00:00 machine-generated
normalization normalization
2026-10-02T18:15:50+00:00 machine-generated
Полная цепочка TPME
[
  {
    "_schema": {
      "id": "https://doi.org/10.5281/zenodo.21287085",
      "name": "AI4LAM TPME",
      "version": "1.0"
    },
    "conventions": [
      "diplomatic",
      "https://github.com/selic/bookolot/blob/main/docs/transcription-guide.md"
    ],
    "features": {
      "bbox_space": "deskewed page, see deskew step",
      "line_bbox": true,
      "line_confidence": true,
      "word_bbox": false
    },
    "file_format": "ALTO 4",
    "human_review_level": "machine-generated",
    "media_id": "http://www.moldavica.bnrm.md/biblielmo?l=ro\u0026a=d\u0026c=carti\u0026d=JD3373",
    "media_title": "Instrukctia dlja zemskih bol'nic Sorocskago Uezda",
    "modification_date": "2026-09-29T22:02:25+00:00",
    "originating_media_file": "26.tif",
    "processing": [
      {
        "application_name": "kraken blla",
        "application_parameters": {
          "device": "cuda",
          "model": "blla.mlmodel (встроенная)"
        },
        "application_provider": "Benjamin Kiessling",
        "application_repo": "https://github.com/mittagessen/kraken",
        "application_type": "line segmentation",
        "application_version": "7.1.1"
      },
      {
        "application_name": "kraken",
        "application_parameters": {
          "device": "cuda"
        },
        "application_provider": "Benjamin Kiessling",
        "application_repo": "https://github.com/mittagessen/kraken",
        "application_type": "OCR",
        "application_version": "7.1.1",
        "inference_model": "ppocrv6_medium (sha256:15313b51ace64cbf)"
      }
    ],
    "provider": "Bookolot",
    "transcript_checksum": "sha256:72a0da3fe24be4d5f059980129da3ba9fdbe313e9424f8da7c434dc8b56f6938",
    "transcript_id": "bookolot:JD3373:26:ocr-v1",
    "transcript_language": [
      "ru-petr1708"
    ],
    "type": "transcript"
  },
  {
    "_schema": {
      "id": "https://doi.org/10.5281/zenodo.21287085",
      "name": "AI4LAM TPME",
      "version": "1.0"
    },
    "characteristics_note": "For reading and search only; quote the parent transcript.",
    "human_review_level": "machine-generated",
    "media_id": "http://www.moldavica.bnrm.md/biblielmo?l=ro\u0026a=d\u0026c=carti\u0026d=JD3373",
    "media_title": "Instrukctia dlja zemskih bol'nic Sorocskago Uezda",
    "modification_date": "2026-10-02T18:15:50+00:00",
    "originating_media_file": "26.tif",
    "parent_transcript_id": "bookolot:JD3373:26:ocr-v1",
    "processing": [
      {
        "application_name": "bookolot.rag.normalize",
        "application_parameters": {
          "normalizer_version": 3,
          "script": "ru_prereform"
        },
        "application_provider": "Bookolot",
        "application_repo": "https://github.com/selic/bookolot",
        "application_type": "spelling normalization",
        "application_version": "fb8b81d-dirty"
      }
    ],
    "provider": "Bookolot",
    "transcript_checksum": "sha256:8085feb270cd7768177884ab053d4add32b3ec85922e0c14cc940384a8f934da",
    "transcript_id": "bookolot:JD3373:26:modern-v3",
    "transcript_language": [
      "ru-t-ru-petr1708"
    ],
    "type": "normalization"
  }
]
Скан этой строки 26

уверенно стоит взглянуть проверьте скан Открыть скан целиком

Текст как напечатаносовременная орфографияоба

как напечатано − 4 + современная орфография − 5 +

  1. 0 . 0.19
  2. 1 m 0.45
  3. 2 ב 0.19
  4. 3 . 0.44
  5. 4 AIB 0.49
  6. 5 E 0.83
  7. 6 Аптечнаго имущества и хирургическихъ 0.99
  8. 7 Зремя пріобрѣ 0.94
  9. 8 предмто 0.87
  10. 9 НАЗВАНІЕ ПРЕДМЕТОВЪ 0.93
  11. 10 Цыоран 0.78
  12. 11 Количество 1.00
  13. 12 Словами 0.94