Detail projektu

Identification of the Czech origin of digital music recordings using machine learning

Období řešení: 01.02.2021 — 31.01.2023

Zdroje financování

Evropská unie - Interní grantová soutěž

- plně financující (2021-02-01 - 2023-01-31)

O projektu

The project aims to create a classifier that will be able to differentiate the string quartet recordings of Czech performers from the rest of Europe/world. The key is to find the parameters of digital recordings (e.g. MPEG-7 parameters), which characterize Czech production. Machine learning will be used for the analysis—k-NN and clustering for the parametrization and convolutional neural networks (CNNs) for the classification. The project builds on the previous thesis of the main proposer, where the potential of this research was identified.

Označení

FEKT-K-21-6874

Originální jazyk

angličtina

Řešitelé

Ištvánek Matěj, Ing., Ph.D. - hlavní řešitel
Miklánek Štěpán, Ing., Ph.D. - spoluřešitel

Útvary

Fakulta elektrotechniky a komunikačních technologií
- příjemce (01.02.2021 - 31.01.2023)
Ústav telekomunikací
- spolupříjemce (01.02.2021 - 31.01.2023)

Výsledky

IŠTVÁNEK, M.; MIKLÁNEK, Š. Exploring the Possibilities of Automated Annotation of Classical Music with Abrupt Tempo Changes. In Proceedings II of the 28th Conference STUDENT EEICT 2022 Selected Papers. 1. Brno: Vysoké učení technické v Brně, Fakulta elektrotechniky a komunikačních technologií, 2022. p. 286-290. ISBN: 978-80-214-6030-0.
Detail

IŠTVÁNEK, M.; MIKLÁNEK, Š. Towards Automatic Measure-Wise Feature Extraction Pipeline for Music Performance Analysis. In 45th International Conference on Telecommunications and Signal Processing (TSP). IEEE, 2022. p. 192-195. ISBN: 978-1-6654-6948-7.
Detail

IŠTVÁNEK, M.; MIKLÁNEK, Š.; SPURNÝ, L. Classification of Interpretation Differences in String Quartets Based on the Origin of Performers. Applied Sciences - Basel, 2023, vol. 13, no. 6, p. 1-20. ISSN: 2076-3417.
Detail

ÖZER, Y.; IŠTVÁNEK, M.; ARIFI-MÜLLER, V.; MÜLLER, M. Using Activation Functions for Improving Measure-Level Audio Synchronization. Proceedings of the International Society for Music Information Retrieval Conference (ISMIR). 2022. p. 750-756.
Detail

NOVÁK, V. Proceedings I of the 28th Conference STUDENT EEICT 2022 General papers. Proceedings I of the 28th Conference STUDENT EEICT 2022 General papers. 1. Brno: Brno University of Technology, Faculty of Electrical Engineering and Communication, 2022. ISBN: 978-80-214-6029-4.
Detail

Odpovědnost: Ištvánek Matěj, Ing., Ph.D.