abcm2ps(1) translate ABC music notation to PostScript


abcm2ps [options] [abc-file [file-options]] [...]

abcm2ps [-h]


abcm2ps translates tunes written in the ABC music notation format to customary sheet music scores in PostScript. It is based on abc2ps 1.2.5 and was developed mainly to print Baroque organ scores that have independent voices played on multiple keyboards and a pedal-board. The program has since been extended to support various other notation conventions in use for sheet music.

Options given immediately after the command name apply to the run as a whole; options given after an ABC file name apply to that file.

Formatting parameters can also be set in `format files' and in the ABC files themselves.


Split tunes across page breaks if necessary.
Start a new page if a tune doesn't fit on the current one (default).
Output one tune per page.
Output multiple tunes per page (default).
-a a
Maximal horizontal compression when staff breaks are chosen automatically. Must be between 0 and 1. This corresponds to the `%%maxshrink a' formatting parameter (default: 0.65)
This option inserts reference elements in the PostScript or SVG output.
-B b
Try to typeset b bars on each staff line. This corresponds to the `%%barsperstaff b' formatting parameter.
Don't try to typeset a fixed number of bars on each staff line (default). This corresponds to the `%%barsperstaff 0' formatting parameter.
-b b
Start measure numbering at b.
The continuation symbol is implicitly appended to each music line. This amounts to automatic line breaking.
Automatic line breaking is turned off (default).
-D directory
Search format files in directory.
-d length
Set the vertical interstaff space to length. This corresponds to the `%%staffsep length' formatting parameter (default: 46pt)
Output is generated in EPS format, one file per page.
-e list
Select which tunes from an ABC file to print. list is either a comma-separated list of tune numbers (as per the `X:' header), or a regular expression which will be matched against the tune headers as a whole. The -e option must occur after an ABC file name and applies to that file.
Ranges of tune numbers may be specified like t1-t2; t2 may be omitted which means `all remaining tunes until the end of file'. Note that filtering may cause problems, e.g., with global (non-tune) definitions in the ABC file.
-F file
Read the format file file.fmt.
Do not read the default format file.
Enable flat beams (useful for bagpipe tunes). This corresponds to the `%%flatbeams 1' formatting parameter.
Omit slurs on grace notes. This corresponds to the `%%graceslurs 0' formatting parameter.
Draw slurs on grace notes (default). This corresponds to the `%%graceslurs 1' formatting parameter.
Produce SVG output instead of EPS. In this mode, each tune goes to a separate file whose name is `Outnnn.svg' (see option -O).
If output is directed to stdout (`-O-'), all SVG images are output without an XML header.
Display the current values of the formatting parameters.
Display program usage hints and quit.
-I length
Indent the first line of the tune by length (default: 0). This corresponds to the `%%indent length' formatting parameter.
-i, +i
Insert a red circle around errors in the PostScript output.
Output a measure number every b measures. If b is 0, the measure number appears at the left of each staff. The trailing b causes a box to be drawn around each measure number (default: no measure numbering). This corresponds to the %%measurenb      formatting parameter.
Don't output measure numbers. This corresponds to the `%%measurenb -1' formatting parameter.
Sets the size of the PostScript output buffer in kibibytes. Specifying a higher value permits the generation of big tunes with -E     or -g (default: 64).
Generate landscape output. This corresponds to the `%%landscape 1' formatting parameter.
Generate portrait output (default). This corresponds to the `%%landscape 0' formatting parameter.
Suppress lyrics. See the %%writefields formatting parameter.
Include lyrics (default). See the %%writefields formatting parameter.
-m length
Set the left margin to length (default: 1.8cm). This corresponds to the `%%leftmargin length' formatting parameter.
Number pages according to the mode:
no page numbers
page numbers at top left of page
page numbers at top right of page
page numbers at top left of even-numbered pages, top right of odd-numbered pages
page numbers at top right of even-numbered pages, top left of odd-numbered pages
For compatibility, -N is equivalent to -N2 (default: -N0).
If a header is defined using `%%header', this option is ignored.
Equivalent to -N0 (no page numbering).
Include notes and history from ABC tune `N:' fields. See the %%writefields formatting parameter.
Omit notes and history from ABC tune `N:' fields (default). See the %%writefields formatting parameter.
-O name
Define the output file name. By default, the output file name is `' for PostScript output, `Outnnn.eps' for EPS output (see -E), or `Outnnn.svg' for SVG output (see -g), where nnn   is a sequence number. If this option is given, the output
name will be name (for PostScript output), or name      will replace `Out' in the output file name (for EPS and
SVG output), respectively. If name is `=', the output file name will be the name of the ABC source file with the extension `.ps', `.eps', or `.svg'. If name is `-', the output is written to stdout.
Revert to the default output file name (`' or `Outnnn.eps')
Quiet mode, only error messages will be shown.
Secure mode; disables file inclusion via %%format and %%EPS and disallows direct PostScript injection via %%beginps and %%postscript.
-s scale
Set the page scale factor to scale. Note that the header and footer are not scaled (default: 0.75). This corresponds to the `%%scale scale' formatting parameter.
-Tn[voice], +Tn[voice]
Activate (or deactivate) tabulature drawing. n is the tabulature number as defined in %%tabulature (up to a maximum of 8 tabulatures), and voice is the voice name, full name or subname as found in V:. When this is absent, the option applies to all voices. Up to 4 such options may be given. See also format.txt.
Output version number and quit.
Produce SVG output instead of simple PS. In this mode, each page goes to a separate output file called `Outnnn.svg' (see option -O).
-w length
Adjust the right margin such that the staff width is length (default: none). This corresponds to the `%%staffwidth length' formatting parameter.
Produce XML+SVG output instead of simple PS. The default file name is `Out.xhtml' (see option `-O').
Include the `X:' tune number in the title. See the %%writefields formatting parameter.
Do not include the `X:' tune number in the title (default). See the %%writefields formatting parameter.


Abcm2ps supports a vast number of formatting parameters that govern the appearance of the typeset output. Please refer to the /usr/share/doc/abcm2ps/format.txt file or the formatting parameter documentation on the official abcm2ps web site (link to URL .



Clefs can be given in K: and V: headers. The full syntax is

[clef=]type[line][+8|-8] [middle=pitch]          


The `clef=' can be omitted when the type is a clef name.

type denotes the clef type. It may be:

A note pitch (G, C, or F)
The pitch indicates which clef is meant: G is the treble clef, C the alto clef and F the bass clef. It also gives the name of the note that appears on the clef's line.
A clef name
The available clef names are treble (clef gives the pitch for G), alto or tenor (C), and bass (F)
perc or P
In percussion mode, accidentals change the glyphs used for note heads. By default, sharp notes are drawn as `x' and flat notes as circled `x'. This may be changed by redefining the PostScript functions pshhd and pflhd.
No clef will be displayed.     

The line gives the number of the line within the staff that the base clef will be written on. The default values are 2 for the treble clef, 3 for the alto clef, and 4 for the tenor and bass clefs.

The +8 and -8 options draw an 8 above or below the staff, respectively. `middle=pitch' (or `m=pitch', for short) is an alternative way of defining the line number of the clef: The pitch indicates what note will be displayed on the middle line of the staff.

When no clef is specified, clef changes between bass and treble will be inserted automatically.

Multi-voice typesetting

Multiple voices may be defined within the header or the tune using

V:name definition ... 

where name is a word consisting of letters and digits only (like violin1). In the tune body, the following notes refer to this voice until another V: is encountered. A definition can be one of:

See above
name=name or nm=name
The name will be displayed at the beginning of the first staff. It can contain \n sequences which will force line breaks. If it contains whitespace it must be double-quoted.
subname=name or snm=name
The name will be displayed at the beginning of all staves except for the first. It can contain \n sequences which will force line breaks. If it contains whitespace it must be double-quoted.
The voice goes on the same staff as the previous voice.
up or down
Forces the direction of the stems for the voice.
dyn=up or dyn=down or dyn=auto
Forces positioning of dynamic marks (above or below the staff) or reverts to automatic positioning (the default)
gstem=up or gstem=down or gstem=auto
Forces the direction of the stems of grace notes (always up or always down) or reverts to automatic positioning (the default)
Reverts to automatic positioning of note stems (up or down) (the default)
lyrics=up or lyrics=down or lyrics=auto
Places lyrics above or below the staff or reverts to automatic positioning (the default)
gchord=up       or gchord=down
Places guitar chords above (the default) or below the
Sets the number of lines on the staff in question. (default: 5)
Sets the scale of the associated staff up to 3. (default: 1)    All other definitions are ignored.

By default, each voice goes on its own staff. The `%%staves definition' pseudo-comment can be used to control staff assignment. The definition consists of voice names (from V:) and pairs of parentheses, braces or brackets.

When a voice name is not within a pair of special characters, it goes on a separate staff.
For voice names enclosed in brackets, a bracket is displayed at the beginning of each line that joins the staves of the voices in question.
For voice names enclosed in braces, all the voices go on two staves (keyboard score). There can be at most four voices between a single pair of braces.
For voice names enclosed in parentheses, all the voices appear on a single staff.

The `|' character prevents measure bars from being drawn between two staves.

If `%%staves' occurs in a tune, all the voices not mentioned will not be output at all.

The %%score directive

The %%score directive occurs in the ABC draft 2.0 standard and is similar to the %%staves       specification described earlier. The rules are:


Voice names within parentheses form a "voice group" and go on a single staff. A voice name that is not within parentheses forms its own voice group and goes on a staff by itself.
Voice groups within braces form a "voice block" and are preceded by a big brace in the output. This is especially useful for keyboard music.
Voice groups or voice blocks within brackets form a "voice block" and will be preceded by a big bracket in the output.
If a `|' character occurs between two voice groups or voice blocks, the bar lines in all of the associated staves will be continuous.
A single voice surrounded by two voice groups can be preceded by an asterisk to make it into a `floating' voice. This means that, for each note of the voice, a separate decision is made whether it is printed on the preceding or the following voice group's staff.
Voices that appear in the tune body but not in the %%score directive will not be output at all. If there is no %%score    directive, each voice will be output on its own
A %%score directive inside a tune resets the mechanism so voices can be removed or added.

Voice overlay

You can add notes to a staff without introducing a complete extra voice by using the ampersand (&). A single measure can be split into two voices like


The (&...&...&) construction allows splitting multiple measures:

|!f!(&GG<G|GG F=E| E2  E(_D/E)|_D D  C      D |C4- |C 
     &DC<C|CC_D C|=B,2_B,B,   |_A,A,(G,/A,/)B,|F,4-|F,)zzD=E| 

A double ampersand (&&) will allow overlaying more than two lines of music but this feature has not yet been implemented.


Aligned lyrics under a staff are written as a w: line directly below the staff line. For example:

edc2 edc2| 
w:Three blind mice, three blind mice 

Each word in the w: line (delimited by blanks) is associated with one note, in sequence. The following special symbols modify this behaviour:

Skips one note.
Splits a word into two syllables which are associated with two adjacent notes. A `-' is drawn between them.
Advances to the next bar line
Is output as a space, but unites two words so they appear under a single note.
Draws a thin underscore from the previous note to the next.

To include more than one line of lyrics, use multiple w: lines. To include hyphens without splitting a word over multiple notes, use -.

If a word starts with a digit, this is interpreted as a stanza number and outdented a bit to the left.

Slurs and ties

The direction of slurs and ties may be controlled using the (, and (', and -, and -', constructions.

Microtone pitches

Microtone pitches are indicated by a fraction after an accidental, as in 3/4c. When omitted, the numerator defaultes to 1 and the denominator to 2 (so /c is the same as 1/2c). The numerator and denominator values may not exceed 256. There is built-in support for quarter-tone accidentals (1/2 and 3/2 sharps and flats); for other values, rendering functions must be defined using %%postscript (see features.txt).

EPS inclusion

EPS files may be included inside tunes using the pseudo-comment `%%EPS file'.


This manual page was written by Anselm Lingnau <[email protected]> for the Debian system (but may be used by others). Permission is granted to copy, distribute and/or modify this document as long as its origin is not misrepresented.