Page Content
to Navigation
Machine Learning and Neural Networks for the Perceptually Relevant Analysis of Music
- © NI, TUB
Music contains structural information as well as semantic connotations which are easy to perceive by human listeners, but which are difficult to extract automatically from an acoustic event (and even from the score of a given piece of music). Here we explore new techniques from the machine learning and the mathematical music theory fields with the goal to create semantically meaningful representations from acoustic events and to automatically extract perceptually relevant patterns from music and sound.
Acknowledgement: Research was funded by the EU and by the Technische Universität Berlin.
Selected Publications:
| Citation key | Adiloglu2007 |
|---|---|
| Author | Adiloglu, K. and Obermayer, K. |
| Title of Book | Communications in Computer and Information Science |
| Pages | 67 – 73 |
| Year | 2007 |
| ISBN | 978-3-642-04578-3, 978-3-642-04579-0 |
| ISSN | 1865-0929 |
| DOI | 10.1007/978-3-642-04579-0_7 |
| Volume | 37 |
| Note | First International Conference, MCM 2007 Berlin, Germany, May 18–20, 2007 Revised Selected Papers |
| Editor | Timour Klouche, Thomas Noll |
| Publisher | Springer Berlin Heidelberg |
| Chapter | Mathematics and Computation in Music |
| Abstract | The similarity neighbourhood model is a mathematical model making use of statistical, semiotical and computational approaches to perform melodic analysis of given music pieces. This paper is dedicated to the investigation of topological features and conditions in connection with the model on the one hand and concrete analyses on the other. Therefore, checking the topological features of the model as well as the analysis results is a good practice not only for theoretical but also for practical reasons. The topological features of the similarity neighbourhood model are investigated from a theoretical viewpoint, in order to figure out under which conditions the collection of the results yielded by the model define a topology. These topological features are then tested practically on the two-voice inventions. These investigations and tests have shown that the similarity neighbourhood model defines a topology not for all cases, but depending on the analysed musical piece. |
| Bibtex Type of Publication | Selected:music |