Eine Übersicht aller Sessions/Sitzungen dieser Veranstaltung. Bitte wählen Sie einen Ort oder ein Datum aus, um nur die betreffenden Sitzungen anzuzeigen. Wählen Sie eine Sitzung aus, um zur Detailanzeige zu gelangen.
Bitte beachten Sie, dass sich alle Zeitangaben auf die Zeitzone des Konferenzortes beziehen. Die momentane Konferenzzeit ist: 09. Mai 2025 15:52:33 MESZ
Post-pandemic era, Internationalisierung und Digitalisierung. Herausforderungen der Musiktheorie in aktueller Forschung und Lehre
Präsentationen
Workshop Themen: Post-pandemic era, Internationalisierung und Digitalisierung. Herausforderungen der Musiktheorie in aktueller Forschung und Lehre Stichworte: Künstliche Intelligenz, Generative AI, Machine Learning, Modellierung
AI for Music Theory
Egor Polyakov1, Christoph Finkensiep2
1HMT Leipzig; 2University of Amsterdam
With the recent advancements in AI, many misconceptions have emerged about its capabilities and applications, particularly within the humanities (“Geisteswissenschaften”). There is a common belief that AI can solve a wide array of scientific problems by "just throwing AI at it." Although potential issues have been discussed in other scientific domains, as highlighted by Bewersdorf et al. (2023) and Emmert-Streib (2020), the discussion of music-related applications of AI is mainly centered around audio-based music generation. In contrast, topics like music analysis and symbolic music generation, which are crucial for the music theory domain, seem to be underrepresented and lack a comprehensive overview of the possibilities of AI applications, especially from a practical point of view.
In our workshop, we aim to provide a systematic introduction to both theoretical and practical approaches in symbolic music-related AI implementations. In the first part of the workshop, we will give an overview of the principle of AI and machine learning and discuss their implications in the context of music theory. The second part is a practical introduction into generative modeling of symbolic music. We will mainly focus on applications for creating "Stilkopie," which refers to the imitation of certain musical styles and forms, and explore case studies based on GAN (Goodfellow et al. 2014) and Diffusion (Ho et al. 2020) architectures.
References
Bewersdorff, A., Zhai, X., Roberts, J., & Nerdel, C. (2023). Myths, mis- and preconceptions of artificial intelligence: A review of the literature. Computers and Education: Artificial Intelligence, 4, 100143. https://doi.org/10.1016/j.caeai.2023.100143.
Emmert-Streib, F., Yli-Harja, O., & Dehmer, M. (2020). Artificial Intelligence: A Clarification of Misconceptions, Myths and Desired Status. Frontiers in Artificial Intelligence, 3, 524339. https://doi.org/10.3389/frai.2020.524339.
Ho, J., Jain, A., & Abbeel, P. (2020). Denoising Diffusion Probabilistic Models. arXiv e-prints. https://doi.org/10.48550/arXiv.2006.11239.
Goodfellow, I. J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., & Bengio, Y. (2014). Generative Adversarial Networks. arXiv e-prints. https://doi.org/10.48550/arXiv.1406.2661.
