htr-united.yml

schema: https://htr-united.github.io/schema/2021-10-15/schema.json
title: Caroline Minuscule by Rescribe
url: https://github.com/rescribe/carolineminuscule-groundtruth
project-name: 'Rescribe''

  '
project-website: https://rescribe.xyz/
authors:
- name: White
  surname: Nick
  roles:
  - transcriber
  - project-manager
- name: "Cl\xE9rice"
  surname: Thibault
  roles:
  - aligner
- name: Karaisl
  surname: Antonia
  roles:
  - transcriber
  - project-manager
description: 'This ground truth repository is a work in process; it currently accounts
  for a part of our complete Caroline Minuscule training pool of around 70 manuscripts
  used for our OCRopus Caroline Minuscule model (see ocropus-models repository).

  '
language:
- lat
script:
- Latn
script-type: only-manuscript
time:
  notBefore: '800'
  notAfter: '1199'
hands:
  count: 1-per-file
  precision: exact
license:
- name: CC-BY 4.0
  url: https://creativecommons.org/licenses/by/4.0/
format: Alto-XML
volume:
- metric: characters
  count: 17155
- metric: files
  count: 17
- metric: lines
  count: 457
- metric: regions
  count: 46
transcription-guidelines: "In general this meant deciding between diplomatic transcription\
  \ (i.e. sticking to what it says on the page) and gently modernized features (i.e.\
  \ reinterpreting medieval signs into modern equivalents) with a view to specific\
  \ categories. Read on for a summary of the rules and the respective rationale behind\
  \ them.\nSUMMARY\nPUNCTUATION\n\n    Modern: medieval punctuation is transcribed\
  \ with modern equivalents; punctus elevatus transcribed as semicolon\n\nCAPITALIZATION\n\
  \n    Diplomatic: Original capitalization retained\n\nABBREVIATIONS\n\n    Diplomatic\
  \ where possible: Retain abbreviations and render glyphs as opposed to expanded\
  \ versions where possible\n    \"*\" where original character isn't served: OCRopus\
  \ (at the point in time of transcription) could not handle some of the medieval\
  \ glyphs, even where a Unicode version was present. Abbreviations not in OCRopus\
  \ are uniformly transcribed as \"*\", in the case of a combined character (such\
  \ as a consonant with a macron) as the base character followed by \"*\" (e.g. \"\
  t*\"). The list of accepted characters in OCRopus can be found in this repository,\
  \ and downloaded and used as codec in the OCRopus training process.\n\nSPACING\n\
  \n    Diplomatic: Preserve manuscript spacing, i.e. give diplomatic transcription\n\
  \nNUMBERS\n\n    Diplomatic: retain original version of both Roman and Arabic numerals'"
characters:
  mode: NFD
  members:
  - i
  - e
  - t
  - u
  - a
  - s
  - n
  - o
  - r
  - m
  - c
  - d
  - l
  - p
  - .
  - b
  - q
  - g
  - '*'
  - h
  - ;
  - "\u0303"
  - f
  - x
  - I
  - "\u0304"
  - E
  - N
  - "\u0328"
  - ':'
  - '&'
  - S
  - "\uA751"
  - C
  - A
  - "\u0111"
  - D
  - U
  - T
  - "\uA753"
  - Q
  - v
  - ','
  - O
  - R
  - P
  - L
  - M
  - "\xE6"
  - H
  - F
  - '?'
  - '1'
  - y
  - "\uA75D"
  - "\uA759"
  - V
  - '4'
  - B
  - z
  - '5'
  - X
  - '6'
  - "\uA75B"
  - /
  - ''''
  - '0'
  - '2'
  - '9'
  - K
  - '-'