Odpri menu

Moderna arhivistika 2020, 3 (1) str./pp. 86-97

Jože GLAVIČ
Zgodovinski arhiv Ljubljana, Enota za Dolenjsko, Novo mesto, Slovenija / Historical Archives Ljubljana, Novo mesto, Slovenia

Primer uporabe programa Transkribus in izdelava modela za avtomatsko optično prepoznavanje znakov za poenostavljeno transkribiranje ročno pisane gotice
Use case of the Transkribus Software and the Creation of a Model for Automatic Optical Character Recognition for Simplified Transcription of Written Gothic

(Moderna arhivistika, III., 2020, št. 1, str. 86-97)
https://doi.org/10.54356/MA/2020/1/ECAD5603

Izvleček:
V prispevku je predstavljeno orodje, ki s pravim pristopom olajša proces transkribiranja. Na podlagi praktičnega primera transkribiranja korespondence družine Terpinc, ki jo hrani Zgodovinski arhiv Ljubljana, sta prikazana proces transkribiranja korespondence in izgradnja modela HTR. Izgradnja modela HTR predstavlja osnovo za nadaljnji proces avtomatiziranega transkribiranja večje količine gradiva. Namen članka je prikazati prednosti in izzive uporabe programskih orodij na zgodovinskih virih, hkrati pa umestiti skupno uporabo procesa transkribiranja in tehnoloških orodij v širši koncept razvoja digitalne humanistike na področju arhivistike in obdelave arhivskih virov.

Ključne besede:
Transkribus, gotica, transkripcija, optično prepoznavanje znakov, orodje za transkribiranje

Abstract:
Use case of the Transkribus Software and the Creation of a Model for Automatic Optical Character Recognition for Simplified Transcription of Written Gothic
The article focuses on a tool, which simplifies the process of transcribing. The process of transcribing and creating a HTR model is demonstrated on a case study based on archival records of the Terpinc family, which is kept in the Historical Archives Ljubljana. The construction of the HTR model forms the basis for a further process of automated transcription of a large amount of material. The purpose of the article is to outline the advantages and challenges of using software tools on historical resources, while at the same time establishing how the transcription process and technology tools can be contextualized in the broader concept of digital humanities, the field of archival science and archival resource processing.

Key words:
Transkribus, Gothic script, transcription, optical character recognition, transcription tool