Home  Publications  Teaching  Software  Data  e-mail

David Meredith's Web Site


Thanks for visiting!

Since 1 September 2009 I've been an Associate Professor (Lektor) in the Department of Architecture, Design and Media Technology at Aalborg University in Denmark. I do research in computational music analysis, music information retrieval and mathematical music theory. You can read what I've written on these topics here.

I am project investigator on the "Learning to Create" (Lrn2Cre8) EU FP7 FET STREP project which aims to build systems that learn from examples to create novel, high quality music. I also lead the Music Informatics and Cognition Research Group.

I teach courses on programming and courses on perception at bachelor and masters level on the Medialogy programme at Aalborg University. You can see some of my teaching materials here. My official Aalborg University page is here.

From 5th November 2007 to 31 December 2008, I was head of music information retrieval at Taptu where I spent most of my time developing software in Java. My projects included web crawlers and fast systems for generating appropriate "thumbnail" snippets from video and audio files.

From 1 May 2006 to 31 October 2007, I was a Lecturer in the Department of Computing at Goldsmiths' College, University of London, where I was a member of the Intelligent Sound and Music Systems (ISMS) group within the Centre for Cognition, Computation and Culture. During this period, I taught courses on Software Engineering and Web design (you can download my lectures and other materials here).

From July 2003 to April 2006, I was a Research Fellow on an EPSRC-funded research project (GR/S17253/02) concentrating on the development of algorithms for musical pattern recognition and extraction.

Within this project, I focused primarily on the construction and evaluation of pitch spelling algorithms. A pitch spelling algorithm predicts the pitch names (e.g., C#4, Db4, etc.) of the notes in a musical passage when given only a "piano roll" type representation (e.g., MIDI). My research in this area led to the development of a new pitch spelling algorithm called ps13. I compared ps13 with most of the other pitch spelling algorithms that have been proposed in the literature by running the algorithms on a large test corpus of tonal music. These comparisons suggest that ps13 is the most accurate pitch spelling algorithm currently available. Some papers that I've written on this topic are available here. The full report on this project is available here (6MB, PDF).

From October 1999 to July 2003, I worked on another EPSRC-funded research project (GR/N08049/01). The main goal of this project was to develop algorithms for discovering perceptually significant repetitions in polyphonic music. This project resulted in the development of new algorithms for pattern discovery and matching in multidimensional music representations. Papers providing full details of these algorithms are available here.

You can read overviews of my research interests here and here.

From October 2002 to July 2005, I was a Visiting Lecturer in the Department of Music at City University, London where I taught courses on Music Perception and Cognition to undergraduates and masters-level students. The full texts of these lectures together with other materials for these courses are available here. You can find an overview of my course on Music Perception and Cognition here.

© 2001-2019 by David Meredith. All rights reserved.