Ambraser Heldenbuch: Transcription and Scientific Dataset 
Mario Klarer / Hubert Alisade / Aaron Tratter
Between January 2017 and December 2019, a research group led by Univ.-Prof. Dr. Mario Klarer at the University of Innsbruck worked on the go!digital 2.0 project Ambraser Heldenbuch: Transcription and Scientific Dataset, funded by the Austrian Academy of Sciences. The project’s aim was to transcribe the entire Ambraser Heldenbuch (Vienna, Austrian National Library, Cod. ser. nova 2663) on the occasion of the 500th anniversary of Emperor Maximilian’s death.
At the beginning of the sixteenth century, Emperor Maximilian I commissioned Hans Ried, a scribe from Bolzano, to copy texts for the Ambraser Heldenbuch. See Alisade (2019). The Ambraser Heldenbuch consists of twenty-five important texts of German medieval literature on 5 + 238 parchment folios. Fifteen of these texts have survived solely in the Ambraser Heldenbuch as Early New High German versions, including Erec by Hartmann von Aue, the anonymous Kudrun, and Mauritius von Craûn. Most modern editions of these uniquely transmitted texts are re-translations into standardized Middle High German that therefore do not represent the original textual source. For many years, scholars and editors have been demanding a transcription of the entire Ambraser Heldenbuch that is faithful to the manuscript text. E.g., Leitzmann (1935); Gärtner (2006); Homeyer and Knor (2015).
The project team transcribed the Ambraser Heldenbuch with the software Transkribus. The generated data were saved as XML files, which contain the transcribed text as well as the coordinates of the respective lines in the manuscript. The three text columns of each page of the Ambraser Heldenbuch were labeled as text regions, which were, in turn, divided into lines. The codex’s unique status as the only source for a number of canonical medieval literary works required an allographic transcription that refrained from any kind of standardization or normalization. For example, variants of letters, such as long ‹s› or rounded ‹r›, were maintained in order to be as faithful as possible to Hans Ried’s use of characters and diacritical signs in the Ambraser Heldenbuch.
Another important aspect of the project was the tagging of, for example, verses, stanzas, and caesuras, as well as abbreviations, initials, and rubrications, all of which provide added value to the transcription proper. The meticulous processing of the texts of the Ambraser Heldenbuch paves the way for a variety of potential applications in further research and editorial endeavors. The consistent allographic transcription of the entire Ambraser Heldenbuch will serve as a reference and basis for linguistic investigations of Early New High German and will facilitate editorial projects on individual texts from the Ambraser Heldenbuch.
In early 2022, the transcription of the entire Ambraser Heldenbuch was published as an eleven-volume print and open access edition by De Gruyter (Klarer 2022a), which, in addition to the allographic and the diplomatic transcriptions of the text, also synoptically juxtaposes the manuscript scans on each page. Furthermore, the data set, together with the transcription guidelines, were published on GitLab (Klarer 2022b) and on Zenodo (Klarer 2022c).
In addition to the eleven-volume edition, a web interface (Klarer 2023) was developed to visualize the project’s data set. Like the edition, the web interface allows for a synoptic presentation of the manuscript scan and the respective transcription. The underlying coordinates of the lines and columns of the manuscript scan in the data set enable the user to click on the lines in the image, thus highlighting the line in the transcription, and vice versa. In contrast to the edition, where the allographic transcription follows the manuscript breaks and the diplomatic transcription follows the verse breaks, the web interface makes it possible to choose the type of transcription (allographic or diplomatic transcription) and the type of breaks (manuscript or verse breaks) independently. The web interface also offers a search function that makes it possible to search all the texts of the Ambraser Heldenbuch for entire words or parts of them. The interface is designed to permit the visualization of additional transcriptions in the future, where the text is linked to the manuscript scan.
In sum, the first complete transcription of the Ambraser Heldenbuch is now accessible via a web interface, through an eleven-volume print and open access edition and through the data set itself. The interface permits different modes of text presentation and has an integrated search function, while the edition ensures long-term archiving of the transcription and the data set lays the foundation for future endeavors by scholars and editors.