This repo contains sheet music typeset with the Lilypond typesetter. The music chosen is in favour of cello music (mostly chamber music) since I am very enthusiastic about this instrument ;-).
Olaf. 2009/12/23
I am doing all this type-setting in Unix environments. I tried TrueOS recently, but I am going back to Debian-based Linux distributions, and Fedora - my current favourite. I am a very bare-bones type of guy so my setup is as follows:
- git from the command line for version management (obviously 😉)
- Emacs for editing the sheet music source and compilation
- Lilypond for rendering
- TiMidity for listening to the MIDI file for errors
- the Fluid GM/GS soundfont for MIDI rendering (which has a pretty decent Cello sound with it)
All the formatting I have done is for the full score, and I am not planning to distribute extracts of partial/single voices. However, it is no big deal to create them by yourself, if you have lilypond on your PC 😎.
You find that the beginning of the master file looks something like this:
\version "2.14.2"
\include "deutsch.ly"
#(set-global-staff-size 19)
\header {
title = "Sonata in a-moll"
composer = "Joseph Bodin de Boismortier (1689-1755)"
arranger = "arr.: Jay Lichtmann"
opus = "op. 7 Nr. 4"
enteredby = "cellist (2011-10-08)"
}
You need to tweak the global staff size parameter to get a nicely formatted output once you have removed or commented out the voices that should not be part of your extract.
Sometimes the master file contains page breaks like this one:
introa = { \time 3/4 \tempo "1. Doucement - Andante " 4=92 }
introb = { \break \tempo "2. Courante - Allegro moderato " 4=152 }
introc = { \pageBreak \time 4/4 \tempo "3. Allemande - Gravement " 8=92 }
introd = { \break \time 3/4 \tempo "4a. Menuetto I - Moderato " 4=132 }
introe = { \break \key a \major \tempo "4b. Menuetto II - Meno mosso" }
I think it is valid to separate most every piece with a \break, but
the page breaks are of course suited towards the full score, and you
may need to adjust those for your limited selection of voices.
The staff group is the place to select which voices you want to include in the output:
\include "v1.ily"
\include "v2.ily"
\include "v3.ily"
music = \new StaffGroup <<
\new Staff {
\set Staff.midiInstrument = \minstr
\set Staff.instrumentName = #"Vc 1"
\transpose a a { \va }
}
\new Staff {
\set Staff.midiInstrument = \minstr
\set Staff.instrumentName = #"Vc 2"
\transpose a a { \vb }
}
\new Staff {
\set Staff.midiInstrument = \minstr
\set Staff.instrumentName = #"Vc 3"
\transpose a a { \vc }
}
>>
Let's say you want to create an output where you combine only the 1st and 3rd voice and drop the middle one. This can be done by simply commenting the part of the middle voice (you could also erase the statements completely, if you like):
\include "v1.ily"
\include "v2.ily"
\include "v3.ily"
music = \new StaffGroup <<
\new Staff {
\set Staff.midiInstrument = \minstr
\set Staff.instrumentName = #"Vc 1"
\transpose a a { \va }
}
%{
\new Staff {
\set Staff.midiInstrument = \minstr
\set Staff.instrumentName = #"Vc 2"
\transpose a a { \vb }
}
%}
\new Staff {
\set Staff.midiInstrument = \minstr
\set Staff.instrumentName = #"Vc 3"
\transpose a a { \vc }
}
>>
If you reverse the selection, then this could be achieved the following way like so:
\include "v1.ily"
\include "v2.ily"
\include "v3.ily"
music = \new StaffGroup <<
%{
\new Staff {
\set Staff.midiInstrument = \minstr
\set Staff.instrumentName = #"Vc 1"
\transpose a a { \va }
}
%}
\new Staff {
\set Staff.midiInstrument = \minstr
\set Staff.instrumentName = #"Vc 2"
\transpose a a { \vb }
}
%{
\new Staff {
\set Staff.midiInstrument = \minstr
\set Staff.instrumentName = #"Vc 3"
\transpose a a { \vc }
}
%}
>>
You could also comment the respective \include in the top part, but
that is not really necessary - it might save you a few microseconds in
rendering the layout though 😉.
I have been adding pieces for clarinets with a single cello recently because I happen to do music with three clarinet players quite regularly now. Being the "bassoon with strings attached", I am facing the beauty of transposing instruments 😩 - I am still trying to understand why music history came up with this idea which makes everything more complicated than need be.
You may wonder why the MIDI files still sound correct despite of this (more or less), and I have to confess: I am cheating here! I create the MIDI files without the transpositions and save them, then I do the sheet rendering with correct transposition, and I finally restore my initial MIDI files and put them under version control. I am pretty sure there is an easier and more straight forward way to do this in lilypond, but I have to find out yet...
As I said before, I am using the Fluid soundfont shipped with many Linux distributions to have the sound output rendered really nicely. If you want to share the result with someone else, and have a Unix-oid machine available, then you may want to check out this small script which creates an OGG file with the sound font rendering included:
$ cat -n midi_ogg.sh
1 #!/bin/bash
2
3 timidity "${1}" -Ow --output-file=- |\
4 ffmpeg -i - -acodec libvorbis "${1%.*}.ogg"
TiMidity creates a WAV file which is then passed on and encoded to OGG Format using the ffmpeg tool. With small adjustments, this could also compress the output to e.g. MP3 or M4A format. A similar routine should also work under Windows.
- Another github repository with arrangements for various chamber music groups
- Choir rehearsal files from Cafe Puccini in Denmark
- The Choral Public Domain Library
- The Clarinet Institute of Los Angeles
- International Music Score Library Project
- MuseScore
- NoteWorthy Composer Scriptorium
- Score Exchange
- Archive from Tobias Schölkopf
- Free sheet music from Johan Tufvesson
Please note this list is very fragile, and sites may disappear or provide no sheet music for free any longer ...
What I would like to do in the future - apart from type-setting more sheet music - is to share my personal favourite music that I have tried myself (should be around 1% of all the pieces, maybe even less), and create a list of recommended pieces per ensemble setup to try with nicely rendered music output for better listening. I need to find some web space for storing those MP3, OGG, ... files, and I am planning to link to them from this GitHub repo, because I think mark-down should be a good way to do that (I am a novice user of this).