Conference Agenda

Conflicts​‍​‌‍​‍‌ of public perspectives on AI in music education consist of comparing joy and hate, but the reality reflected in the everyday classroom practice is stable and quite complicated. This article is about how the music education sector faces AI, adjusts to it, and limits its expansion in music teaching, away from the theoretical discussions to the practical sides of schools.

Using the findings of the music education research, the authors show how teachers work with digital music resources, AI-driven applications, and school library materials under conditions existing in the outside world such as limited time, data privacy, curriculum standards, and institutional requirements. The research does not position AI as a problem or a solution but rather as a stage-set integration process which is variable and dependent on the context.

The article presents a flexible model of AI adoption in music education that considers pedagogical goals, moral issues, and infrastructural challenges based on these facts. The transformation of music libraries and public institutions as the intermediaries between the technological and educational spheres gets a significant part of the paper's attention. Collaborative methods such as curated digital collections, annotated scores, and guided access to historical materials are talked over as possible but disputed ways of improving student engagement.

The paper, referring the local research of South Korea and Germany, demonstrates the differences in strategies of traditions, technology, and access negotiation. It concludes by reflecting on the realistic role of music libraries in supporting music education amid technological and societal change.

Within the field of artificial intelligence (AI), there are so-called Large Language Models (LLMs) that can generate human-like content, synthesize information, translate, explain concepts, and interact with users in ways that help them gain insights into a topic. Writing with AI tools raises ethical questions and issues related to copyright and academic integrity policies. Should there be boundaries for using LLMs in academic writing? Or how much is too much? How can AI be used to gain insight without slipping into imitation and losing one’s personal voice? What does authenticity mean in co-writing with AI? And how can librarians evaluate AI-assisted texts (articles, books) in order to decide whether they should be included in a library’s acquisition list? This paper focuses on ethical issues regarding whether and how AI should be used in academic writing. It addresses these questions in an interactive manner, supported by reference sources aligned with the policies of the European Union and UNESCO concerning the ethical use of artificial intelligence in research and academic writing.

This paper presents a computational study of the National Taiwan Symphony Orchestra (NTSO) Archive to examine how the aesthetic vocabulary surrounding Western classical music has evolved in Taiwan over the past six decades. As Taiwan’s oldest symphony orchestra, the NTSO provides a vantage for understanding the formation of a local canon, the development of listening practices, and the shaping of aesthetic values within changing cultural contexts. By analysing long-term programming history, the study highlights the orchestra’s role as a structural force in the development of Taiwan’s contemporary musical identity.

The project draws on a digitised archive of approximately 1,300 programme booklets from 1945 to 2025, documenting more than 2,500 performed works and over 2,100 musicians. These booklets serve as rich artefacts of reception history, revealing how the orchestra has interpreted Western repertoire, framed aesthetic concepts, and mediated musical meaning for the public.

Using an interdisciplinary framework that integrates musicology, digital humanities, and advanced AI-based textual analytics, the study applies large language models (LLMs), text mining, and topic modelling to trace the evolution of aesthetic terminology and long-term patterns in repertoire selection. AI-driven methods allow precise, large-scale analysis of shifting interpretive language and stylistic emphases across different periods.

Preliminary findings show increasing repertoire diversity and a transition from early, formulaic descriptors to more nuanced and locally inflected interpretive language. By uncovering quantitative and linguistic trends, this study illuminates how Taiwan has imagined and articulated its classical-music aesthetic system, while demonstrating the value of applying state-of-the-art AI tools to music library archives.