045 Digital Research Infrastructure for the Arts and Humanities (DARIAH-PL) at Adam Mickiewicz Uniwersity

Authors: Brygida Sawicka-Stępińska, Katarzyna Klessa

Adam Mickiewicz University in Poznań is one of the partners of DARIAH-PL – the largest consortium of humanities in Poland consisting part of the pan-European infrastructure DARIAH-EU. Since 2020, the consortium has carried out two projects – Dariah.lab (2021-2023) and Dariah-hub (2024-2025) – with an objective to support and expand research infrastructure of digital humanities both in the purely scientific context, as well as in the area of practical applications in the economy.

The outcome of Dariah.lab was a network of dispersed laboratories and resources. Dariah-hub aims to develop new state-of-the-art functionalities that build upon this foundation and integrate them through a common digital platform, providing tools for interdisciplinary research on Polish heritage and architecture to a wide audience.

Adam Mickiewicz University in Poznań hosts six modules of DARIAH infrastructure. The modules include:

• MultiCo (at the Faculty of Modern Languages and Literatures): a multimodal corpus and recording studio with Motion Capture infrastructure. The resources include a collection of audio and video recordings with multilayer descriptions of speech, gesture and body movement across different speaking styles: monologs, dialogs or multi voice discussions. Current work includes developing software for automated analysis and visualization of gesture space, rhythm and interspeaker accommodation based on Biovision Hierarchy (BVH) data derived from Motion Capture output.
• Philological Hub (at the Faculty of Polish and Classical Philology): a module aimed at developing tools for intelligent literary and linguistic research, such as deep source analysis, mapping and visualizing literary works, or creating and researching electronic literature, as well as an archival source acquisition and digitization (e.g. 10,000 pages of “Kaliszanin”, a periodical published in the 19th century, previously available only via St. Petersburg library of Russian censorship, now deposited in a publicly available scientific repository). Current works include construction of archives of Polish regional language varieties and Polish electronic literature.
• HANOI (at the Faculty of English): tools and resources for the automatic analysis of handwritten notes of conference interpreters based on algorithms which scan and analyze handwriting, convert it to pixels and calculate selected parameters. Current work includes adding a multimodal component to handwritten resources and the development of infrastructure supporting multimodal analysis of interpreters’ verbal and non-verbal communication as well as the role of note-taking in memory and information recall.
• Audiovisual Archives of the Faculty of Art Studies: art reproduction tools and infrastructure as well as unique digitized collections including about 40,000 items (reproductions of works of art), as well as documents and audio recordings. The online repository Audiovisual Archives features extensive record functionality, support for “hashtags”, administrator-definable list of meta-information with support for dictionaries and access management. The Archive offers a fee-based service for digitizing audio information from analogue media. Current works include further digitization of archival audiovisual resources.
• LinguoMusa (at the Faculty of the Art Studies): a laboratory geared towards a comparative analysis of two types of human communication: natural languages (speech and gesture) and music (singing and instrumental music). The laboratory features a database of music reflecting the contemporary music environment in Poland. Current work includes the analysis of pitch and rhythm distribution on a large dataset of music samples, as well as construction of an interactive graph visualizing diachronic development of human music and natural speech.
• Tools for text normalization and diachronic analysis (at the Faculty of Mathematics and Computer Science): the laboratory provides tools for diachronic text analysis used to date and normalize texts, such as software for automatic diachronic normalization of Polish texts that makes historical texts more “contemporary” by translating them to contemporary language, software for text dating with an average accuracy of 14 years, software for replacing words with contemporary synonyms, or software for searching texts for the same persons referred to using different names. Current works include enhancement of the OCR tools using the resources provided by other AMU DARIAH-PL resources (HANOI and “Kaliszanin”).