PatternViewer

Designed by Ali Nikrang, Tom Collins, and Gerhard Widmer, the PatternViewer application plays an audio file of a piece synchronized to a point-set representation, where the colour of the points represents an estimate of the local key. The pendular graph in the top-left corner represents the piece’s repetitive structure, and can be clicked to find out about motifs, themes, and other repetitive elements.

Below you can download version 1.0 of the PatternViewer, as well as music with which to try it.

Download/install for Mac OSX

1. Download this zip file by clicking on the link.
2. Unzip the file by double-clicking on it.
3. Move the unzipped ‘PatternViewer_ver1_0’ folder to an appropriate location, such as Applications (you may be prompted to enter your machine password).
4. Having downloaded some music with which to use the PatternViewer (see section below on ‘Downloading music’), navigate to the ‘PatternViewer_ver1_0’ folder and, inside this folder, double-click on the PatternViewer icon.
5. A blank PatternViewer interface should appear, and, in front, a dialogue window for locating and opening a piece of music. Use the dialogue window to locate the folder for the piece of music you would like to open, and, within this folder, select a file along the lines of config*.txt, either by double-clicking on config*.txt or single-clicking on ‘Open’.
6. The piece of music should open, with accompanying visualisations, and begin to play. For more details on using/understanding the PatternViewer interface, please see the readme.txt file in the ‘PatternViewer_ver1_0’ folder.

Download/install for Linux

1. It is possible to use PatternViewer on almost every popular Linux desktop distribution with Wine Compatibility Layer. It can be installed on numerous operating systems, and in most cases is available in the official repositories for that system. It is a rather large package (~650 MB, sorry!), and can be obtained in one of two ways:
   • If you are running Ubuntu, open up the Ubuntu Software Center, type ‘wine’, and then install ‘wine, Windows Compatibility Layer (meta-package)’. There are also some other utilities available for wine, but they are not required for running the PatternViewer;
   • If you prefer to do things from the command line, open up a Terminal window, type
     sudo apt-get install wine
     hit return and follow any install instructions that appear in the Terminal.
2. Download this zip file by clicking on the link.
3. Unzip the file using Ubuntu's file extractor, or by opening up a Terminal window, navigating to the location of the downloaded zip file, typing
   unzip "/blah/PatternViewer_ver1_0.zip"
   and hitting return.
4. Move the unzipped ‘PatternViewer_ver1_0’ folder to an appropriate location, such as Applications (you may be prompted to enter your machine password).
5. After installing Wine, unzipping the above file, and downloading some music with which to use the PatternViewer (see section below on ‘Downloading music’), open a Terminal window and navigate to the directory in which you unzipped ‘PatternViewer_ver1_0’.
6. Run PatternViewer.exe from there by typing
   wine PatternViewer.exe
   Alternatively you can run PatternViewer without using any command lines; just right-click on PatternViewer.exe and select ‘Open With Wine Windows Program Loader’.
7. A blank PatternViewer interface should appear, and, in front, a dialogue window for locating and opening a piece of music. Use the dialogue window to locate the folder for the piece of music you would like to open, and, within this folder, select a file along the lines of config*.txt, either by double-clicking on config*.txt or single-clicking on ‘Open’.
8. The piece of music should open, with accompanying visualisations, and begin to play. For more details on using/understanding the PatternViewer interface, please see the readme.txt file in the ‘PatternViewer_ver1_0’ folder.

Download/install for Windows

1. Download this zip file by clicking on the link.
2. Unzip the file by double-clicking on it.
3. Move the unzipped ‘PatternViewer_ver1_0’ folder to an appropriate location, such as Program Files.
4. Having downloaded some music with which to use the PatternViewer (see section below on ‘Downloading music’), navigate to the ‘PatternViewer_ver1_0’ folder and, inside this folder, double-click on the PatternViewer icon (PatternViewer.exe).
5. A blank PatternViewer interface should appear, and, in front, a dialogue window for locating and opening a piece of music. Use the dialogue window to locate the folder for the piece of music you would like to open, and, within this folder, select a file along the lines of config*.txt, either by double-clicking on config*.txt or single-clicking on ‘Open’.
6. The piece of music should open, with accompanying visualisations, and begin to play. For more details on using/understanding the PatternViewer interface, please see the readme.txt file in the ‘PatternViewer_ver1_0’ folder.

Downloading music

• First movement of Symphony no.1 in C major op.21 by Ludwig van Beethoven (1770-1827), in a performance by Walter Weller and City of Birmingham Symphony Orchestra.
• First movement of Symphony no.3 in E-flat major op.55 by Ludwig van Beethoven (1770-1827), in a performance by Serge Koussevitzky and the Boston Symphony Orchestra. (First seven minutes only; there is a mistake in the symbolic encoding that will take some time to correct.)
• ‘The Silver Swanne’ by Orlando Gibbons (1583-1625), in a piano performance of the original vocal score by Tom Collins.

A word on the music data

We have included more symbolic representations of the above pieces than is strictly necessary. This is because we want future researchers to be able to replicate the process of taking a high-quality symbolic representation (such as kern) and automatically generating the audio-synchronised point-set representations on which PatternViewer is based. For interested researchers, replication can be achieved by downloading and installing MCStylistc, and then running the Common Lisp scripts converter.lisp and estimate-audio-time.lisp enclosed with each of the above pieces of music.

Given moderate funding, we are optimistic about how quickly the above list of music data might grow. Many of the processes involved in producing synchronised audio-symbolic representations are on their way towards being automated. It would be a great project to bring together open-source algorithms for these various processes, and to use them to produce a large amount of high-quality synchronised audio-symbolic data in a short period of time.

Questions/comments?

Please feel free to contact Ali Nikrang (an(at)musicresearch.eu) or Tom Collins (tom.collins(at)dmu.ac.uk) with questions or comments about the PatternViewer.