Bookolot deep dive into books

page 4

4 / 222 Moldovan Cyrillic confidence 0.98 not checked by a human

Provenance

transcript transcript
2026-09-17T21:29:38+00:00 machine-generated
transliteration transliteration
2026-09-18T07:36:04+00:00 machine-generated
Full TPME chain
[
  {
    "_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=JD4398",
    "media_title": "Gramatica limbii moldoveneşti. P. 1 : Fonetica şi morfologia",
    "modification_date": "2026-09-17T21:29:38+00:00",
    "originating_media_file": "4.tif",
    "processing": [
      {
        "application_name": "kraken blla",
        "application_parameters": {
          "device": "cpu",
          "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": "cpu"
        },
        "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:e0ed695b40e8fd919eae24de49f222c04c879ac104eef298af37159832e793ed",
    "transcript_id": "bookolot:JD4398:4:ocr-v1",
    "transcript_language": [
      "ro-Cyrl-MD"
    ],
    "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=JD4398",
    "media_title": "Gramatica limbii moldoveneşti. P. 1 : Fonetica şi morfologia",
    "modification_date": "2026-09-18T07:36:04+00:00",
    "originating_media_file": "4.tif",
    "parent_transcript_id": "bookolot:JD4398:4:ocr-v1",
    "processing": [
      {
        "application_name": "bookolot.rag.normalize",
        "application_parameters": {
          "normalizer_version": 3,
          "script": "md_cyrillic"
        },
        "application_provider": "Bookolot",
        "application_repo": "https://github.com/selic/bookolot",
        "application_type": "transliteration",
        "application_version": "851f6b2-dirty"
      }
    ],
    "provider": "Bookolot",
    "transcript_checksum": "sha256:aa99da9fccbd58da698f8ee00ea1c87ec2cd451cfaa8715972851a2dac459cba",
    "transcript_id": "bookolot:JD4398:4:modern-v3",
    "transcript_language": [
      "ro-Latn-t-ro-cyrl-md"
    ],
    "type": "transliteration"
  }
]
Scan of this line 4

confident worth a look check the scan Open the full scan

Text as printedtransliterationboth

as printed − 2 + transliteration − 2 +

  1. 0 Синтакса, орь штиинца деспре алкэтуире (гречеште син¬ 0.98
  2. 1 таксис— алкэтуире, композицие), ынвацэ кипуриле алкэ¬ 0.98
  3. 2 туирий пропозицией. 0.98
  4. 3 Ортография ынвацэ а скри дрепт кувинтеле, яр ортое¬ 0.98
  5. 4 пия — а рости дрепт кувинтеле. Пунктуация есте штиинца 0.97
  6. 5 деспре фолосиря дряптэ а сэмнелор де пунктуацие. 0.98