HTR Winter School 2025 - Late Medieval Latin

Description: This dataset contains ground truth for late-medieval Latin scripts, based on selected Central European manuscripts dating from the 13th to 15th centuries. It was created by participants of the Late Medieval Latin group during the Winter School: HTR of Historical Sources, held in Vienna in 2025. ⚠️ With the exception of folios 68v–75v of the manuscript Vienna, Austrian National Library (Österreichische Nationalbibliothek), Cod. 508 - 13th c., all transcriptions have been reviewed at least once. However, please be aware that they may still contain errors. All transcriptions are available also on the GitHub repository: https://github.com/HTR-School-Vienna/2025--late-medieval-latin 🌐 You can browse transcriptions directly on GitHub Pages: https://htr-school-vienna.github.io/2025--late-medieval-latin/ Dataset: Přibík of Pulkava’s Chronicle 02 - Praha, Národní knihovna ČR, sign. I C 24 - second half of the 15th c. 🔗 Digital images of the manuscript: https://www.manuscriptorium.com/hub/catalog/default/detail/single/manuscriptorium%7CAIPDIG-NKCR__I_C_24______2MRQ2U9-cs?lang=cs ✍️ Transcription: Leila Leoni: 38v-45r 06 - Praha, Knihovna Národního muzea, sign. VIII F 49 („Stehlíkovský“) - first half of the 15th c. 🔗 Digital images of the manuscript: http://www.manuscriptorium.com/apps/index.php?direct=record&pid=AIPDIG-NMP___VIII_F_49___2Z378T3-cs ✍️ Transcription: Hagar Barak: 7v-15r Olga Kalashnikova: 15v-22v Jennifer Kostoff-Kaard: 23v-30v Angelica Coppola: 38v-44r 08 - Praha, Národní knihovna ČR, sign. XIX B 5 - 1380-1500 🔗 Digital images of the manuscript: http://www.manuscriptorium.com/apps/index.php?direct=record&pid=AIPDIG-NKCR__XIX_B_5_____1OYMOQD-cs ✍️ Transcription: Urszula Zachara-Związek: XXIIIIr-XXXIr 17 - Munich, Bavarian State Library (Bayerische Staatsbibliothek), sign. Clm. 476 - circa 1476 🔗 Digital images of the manuscript: https://www.digitale-sammlungen.de/en/view/bsb00076444?page=,1 ✍️ Transcription: Aliaksandra Valodzina: 18v-24v Fabian Andre: 31r-38r 19 - Prague, National Library of the Czech Republic (Národní knihovna ČR), sign. I D 10 - second half of the 15th c. 🔗 Digital images of the manuscript:http://www.manuscriptorium.com/apps/index.php?direct=record&pid=AIPDIG-NKCR__I_D_10______0NXXRKA-cs ✍️ Transcription: Agnieszka Ziemińska: 127v-131r Aeneas Silvius Piccolomini: Historia bohemica Vienna, Austrian National Library (Österreichische Nationalbibliothek), Cod. 3445 - circa 1485-1496 🔗 Digital images of the manuscript: https://viewer.onb.ac.at/10264119/9 ✍️ Transcription: Federico Rossi: 35r-42r Petr Šámal: 42v-49v Denise Ugliano: 50r-57r Noam Lefler (not reviewed): 68v-75v Cosmas of Prague Vienna, Austrian National Library (Österreichische Nationalbibliothek), Cod. 508 - 13th c. 🔗 Digital images of the manuscript: https://viewer.onb.ac.at/13227113 ✍️ Transcription: Thomas Paul Warren: 1r-6v Amelie Paulsen: 17r-24r Neal Bold: 7r-9r, 26r-30v Michał Tadeusz Noworyta: 9v-16v Dalimil`s Chronicle Latin Fragment - 14th c., Prague, National Library of the Czech Republic (Národní knihovna ČR), XII E 17 🔗 Digital images of the manuscript: https://www.manuscriptorium.com/hub/catalog/default/detail/single/manuscriptorium%7CRASTIS-NKCR__XII_E_17____481A7Q1-cs?lang=cs ✍️ Transcription: Riccardo Cassi: 1r, 2r, 5r-v, 7r-9v, 10v-12r Transcription guidelines: For the transcription guidelines, see the corresponding file. Data organisation: The data are organised in eight folders, one for each manuscript. Each folder contains two subfolders: 'Ground Truth' and 'Done'. The 'Ground Truth’ folder contains transcribed pages that the organisers have checked, while the 'Done' folder contains finished pages that have not been checked. The transcriptions are in the form of XML files in the format of PageXML, with one file per page. Each manuscript folder also contains an image folder, which contains the images of the manuscript pages. The images are in the form of JPEG files, with one file per page. The images are named according to the folio number of the manuscript page. How to cite: This dataset was created by: Jan Odstrčilík, Zuzana Čermáková Lukšová, Annamária Kovács, Fabian Andre, Hagar Barak, Neal Bold, Riccardo Cassi, Angelica Coppola, Olga Kalashnikova, Jennifer Kostoff-Kaard, Noam Lefler, Leila Leoni, Michał Tadeusz Noworyta, Amelie Paulsen, Federico Rossi, Petr Šámal, Denise Ugliano, Aliaksandra Valodzina, Thomas Paul Warren, Urszula Zachara-Związek, Agnieszka Ziemińska. The digitisation is not copyright free, but the transcription is. However, properly annotating a corpus takes time and is a task that should be recognised. If you use any item from this corpus as ground truth, cite the dataset using the following information: