Skip to content

openmusic-project/RQ

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

21 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

RQ

Supervised approach of rhythm quantification based on tree series enumeration

The RQ library for OM proposes a new interactive approach for rhythm quantification. In contrast with previous single-solution systems, RQ runs a graphical interface that allows to visualize and edit the transcriptions of the segmented input sequence.

A. Ycart, F. Jacquemard, J. Bresson — UMR STMS Ircam/CNRS/SU, 2015-2016

The k-best parsing

The quantization algorithm is a recursive, dynamic-programming, lazy algorithm. It works by recursively subdividing segments into equal parts, and then aligning the input points to the closest segment boundary. The algorithm enumerates the subdivisions that give the best results, ranked according to a criterion that combines two quality measures :

  • Distance: how close the output is to the input. This distance is given by the sum of the differences between each input point and the point to which it is aligned.
  • Complexity: in other words, readability of the output. Complexity takes into account the size of the output rhythm tree, the arities of the tree nodes, and the number of grace notes in the final notation.

Compatibility

OpenMusic 6.10 and higher.

Documentation

See Project Wiki

Further reading

The RQ transcription system is described in:

More in-depth description of the algorithm can be found in:

See also a presentation of the library (in French).

The algorithm is based on the algorithms described in:

  • L. Huang, D. Chiang, Better k-best parsing, Proceedings of the Ninth International Workshop on Parsing Technology. Association for Computational Linguistics.

The segmentation algorithm is described in:

More on rhythmic notation, rhythmic quantification and rhythm trees:

For any inquiries, please contact : adrien[dot]ycart[at]ircam.fr