On Fri, 25 May 2012 17:28:02 +0000 (UTC)
Klaus Föhl <[email protected]> wrote:

> Hello,
> 
> I like the lilypond notation using \relative being concise and readable.
> Entering on a computer keyboard is fairly quick, but still it feels
> that playing a melody line would be so much quicker. In particular
> if one does not have a typing c4 d e f g1 style but c4 d4. e8 f8. g16 c,1
> 
> What "better" methods exist?
> 
> For example I have looked into rosegarden output. 
> Minor issue:the output is not in relative notation.
> More cumbersome are slightly non-aligned notes to the beat
> (me being an imperfect human) and in particular varying
> note lengths introducing rests where the music and the audible sound
> both have none.
> 
> I have seen techniques where the pitch is via piano keyboard
> and rhythm is via computer keyboard. I am not fully convinced.
> 
> I have seen a custom-designed computer keyboard that combines
> pitch and duration. It might work well after a learning curve.
> 
> What I am tempted is to take midi file information (i.e. gunzip a.rg),
> or the rosegarden ly output and reverse-engineer it into event lists.
> Whatever the detail: only piano-keyboard input and get both pitch and length.
> 
> Then to apply some smart quantisation. For one thing notes like c1
> are much more likely than c2... or alignment with bars is probable,
> aspects that require some adaptive rules, possibly some parameter training.
> Also the routine should pick up and follow the meter as played,
> as opposed to techniques providing the rigid mentronome frame.
> 
> Well, before I reinvent the wheel myself: are such things already out there?
> 
> Cheers
> Klaus
>  

There are some GUIs that make it easier to use actual notes to create Lilypond 
files.
How well they perform.. I'll leave this discussion to other people.

Nils

_______________________________________________
lilypond-user mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/lilypond-user

Reply via email to