LilyPond: anacrusis, tuplets, multiple voices
Scriabin, Sonata op. 19Table of Contents
A very short introduction
LilyPond is an incredible and powerful language for writing music scores.
Every musician should learn it. But writing code is more suitable for programmers than for musicians and many attempts have to be made to write complete scores.
In this post I show you some coding snippets regarding anacrusis, tuplets and multiple voices (If you’re reading this article you know what they are).
Here is the basic patterns for each of them in LilyPond language:
| Musical item | LilyPond pattern |
|---|---|
| Anacrusis | \partial x followed by a x duration note |
| Tuplets | \tuplet x/y {n1 n2 n3}: n(1, 2, 3) are the notes with “x” total duration instead of “y” |
| Multiple voices | << {musical expression 1} \\ {musical expession 2}, that’s the double backslash construct |
Colored items in a real score
In the image shown at the top of the page you can see a fragment of a real score, limited to the first two measures, from the Piano Sonata No. 2 op. 19, Sonate-Fantaisie, by Alexander Scriabin (Aleksandr Nikolaevič Skrjabin).
Colors identify musical elements as follows:
| Musical item | Color |
|---|---|
| Anacrusis | Red |
| Tuplets | Blue |
| Multiple Voices | Blue and Magenta |
I used the color blue to highlight both the tuplets and the outer voice (magenta for inner voice).
Analysis of elements
The anacrusis, in this example, is realized by two simultaneous notes with the same duration.
The pattern for contextual notes having the same duration, such as octaves or chords, is the following < ... >.
So the fragment ( without dynamic indications) for right and left hand, including simultaneous notes, is as follows:
right = \relative c'' { \partial 16 < dis,, gis dis' >16 }
left = \relative c' { \partial 16 < gis,, gis' >16 }
The tuplets (triplets) are, in this score, a little more complicated: double notes (colored blue) and separate inner voice (colored magenta).
Here is the code for the tuplet only, without the inner voice, from the first measure:
\tuplet 3/2 { < gis, gis'>8 < gis gis' > < gis gis' > }
Now the code for the tuplet only, without the inner voice, from the second measure:
\tuplet 3/2 { < fisis, fisis'>8 < fisis fisis' > < fisis fisis' > }
Now the multiple voices.
Here is the code for the tuplet and the inner voice from the first measure:
{
<<
\tuplet 3/2 { < gis, gis'>8 < gis gis' > < gis gis' > }
\\
< dis' eis >4
>>
}
And finally, the code for the tuplet and the inner voice from the second measure:
{
<<
\tuplet 3/2 { < fisis, fisis'>8 < fisis fisis' > < fisis fisis' > }
\\
dis'4
>>
}
}
Thank you for your attention.
