Jump to content

Sonic Visualiser

From Wikipedia, the free encyclopedia
Sonic Visualiser
Developer(s)Centre for Digital Music at Queen Mary, University of London
Stable release
5.0 / 31 July 2024; 3 months ago (31 July 2024)
Repository
Written inC++
Operating systemLinux, MacOS, Windows
TypeAudio analysis
LicenseGPL-2.0-or-later[1]
Websitewww.sonicvisualiser.org

Sonic Visualiser is an application for viewing and analysing the contents of music audio files. It is a free software distributed under the GPL-2.0-or-later licence.[2]

History

[edit]

Sonic Visualiser was developed at the Queen Mary University of London's Centre for Digital Music in 2007. It was written in C++ with Qt and released under the terms of the GNU GPL.[2]

Overview

[edit]
Screenshot of the spectrum of the refrain of a pop song (precisely "Più bella cosa" by Eros Ramazzotti): basses, drums and artist's voice can clearly be identified.
Sonic visualiser melodic range spectrogram example

Sonic Visualiser represents acoustic features of the audio file either as a waveform or as a spectrogram.[3] Spectrogram is a heatmap, where horizontal axis represents time, vertical axis represents frequency, and the colors show presence of frequencies. Sharpness and smoothness of the spectrogram can be configured.[4] There are three types of spectrogram:

  • generic spectrogram
  • melodic-range spectrogram
  • peak-frequency spectrogram

Generic spectrogram covers the full frequency range and uses linear frequency scale. Melodic-range spectrogram covers the range which usually contains musical detail. Peak-frequency spectrogram performs phase difference calculations and estimates exact frequencies at each peak cell.[2]

The interface consists of panes and layers. Panes allow to display multiple visualisations simultaneously, and they get aligned in the time axis. A pane can have multiple layers which are used for annotation.[2] The user can configure color schemes for layers, and they can be navigated by clicking the labeled tabs.[4]

There are multiple types of annotation layers which can be edited, including time instants, time-value plots, labels and images. Time instants do not have any associated value, and they can be used to annotate points (e.g. beat locations).[2] Annotations allow the user to clarify relationships between musical parameters.[3]

Sonic Visualiser supports third-party plugins in the Vamp plugin format. The plugins take audio input and parameters and return values for display.[2] There are plugins which compute spectral flux and spectral centroid. Other plugins include automatic melody extraction, beat finding, chord analysis, etc.[4]

Sonic Visualiser is available for Linux, OS X, and Windows operating systems.[2]

See also

[edit]

References

[edit]
  1. ^ "Sonic Visualiser license". Archived from the original on 2021-05-09.
  2. ^ a b c d e f g Cannam, Chris; Landone, Christian; Sandler, Mark (October 2010). "Sonic visualiser: An open source application for viewing, analysing, and annotating music audio files" (PDF). Proceedings of the 18th ACM international conference on Multimedia. pp. 1467–1468. doi:10.1145/1873951.1874248. ISBN 978-1-60558-933-6. Retrieved January 18, 2024.
  3. ^ a b Dolan, David; Slaboda, John; Jensen, Henrik Jeldtoft; Cruts, Bjorn; Feygelson, Eugene (2013). "The improvisatory approach to classical music performance: an empirical investigation into its characteristics and impact" (PDF). Music Performance Research. 6. Retrieved January 19, 2024.
  4. ^ a b c Thompson, Marc R. (December 1, 2021). "Sonic Visualiser: Visualisation, Analysis, and Annotation of Music Audio Recordings". Journal of the American Musicological Society. 74 (3): 701–714. doi:10.1525/jams.2021.74.3.701. Retrieved January 20, 2024.
[edit]