Skip to main content
Topic: Real Music Composition Software (Read 8524 times) previous topic - next topic

Real Music Composition Software

I've been studying how Expert Systems mimic pro composers. I'd really like to see some music composition software which isn't only  music sequencing software or music notation/graphing software.

People listen to other people's music and copy it while simultaneously transforming it to make it new and unrecognizable as the original. Eventually this process turns into the composers own unique voice.

I'd specifically like to see a program that will give novices the same transformation tools that pros use to compose; such as take a melody I input, perhaps only scanned in, that the computer then runs through all of the traditional music composition transformational devices without making me write them all out and test them on my instrument before I input them into the so called music composition software.

If and when it can instantly create, inversions, retrogrades, retrograde-inversions, permutations, augmentations, diminutions, diatonic transpositions, INTERVERSIONS, putting the original melodic series together with the rhythmic retrograde, or putting the melodic series' retrograde inversion with the original melodic rhythm or taking the melodic series from one tune and combining it with the melodic rhythm from another tune, automatically make a diatonic melody pentatonic or automatically change the mode, then and ONLY then will there be any real music composition software in the universe.

Re: Real Music Composition Software

Reply #1
Actually, Gongchime, that's been done for at least fifty years. I have a recording of something called the Illiac Suite for String Quartet, composed in 1957 by an Illiac computer at the University of Illinois programmed by Lejaren Hiller and Leonard Isaacson. Parts of it are interesting, parts of it are beautiful, but it is all lifeless. The things you describe that a computer can do are all techniques for making music; they aren't music. It would be nice if the computer could produce retrogrades, inversions, etc. for the composer, but it will never be able to figure out where to put them for the best musical effect: that takes a human ear.

Regards,

Bill

Re: Real Music Composition Software

Reply #2
Well said Bill,
Gongchime, what you are requesting takes all the creative input of the composer out of the equation.  I take your point from a literal perspective, but I would never use such a product - I may not be much of a composer, but what I create is mine - not the product of some soulless machine.  JMHO
I plays 'Bones, crumpets, coronets, floosgals, youfonymums 'n tubies.

Re: Real Music Composition Software

Reply #3
Ditto, William and Lawrie.  Might as well have software to compose novels and poems.  In fact, Roald Dahl wrote a short story with just that premise, something like "The Amazing ???????? Machine"  Rather chilling ending.  Shows how we can lose our humanity when we make imagination and creativity into commodities to be sold like so many grain futures or the like.

Re: Real Music Composition Software

Reply #4
There is also a very funny book called The Silver Eggheads by Fritz Leiber. Same basic premise, but with a comic twist.

 

Re: Real Music Composition Software

Reply #5
I always feel somewhat guilty when I don't write things out on manuscript paper first.  And, I'd never use a midi input.  The closest I'm inclined to get is to record something on my flute, turn it into a midi (have software for this) and then go to Noteworthy & clean it up.  I'd only do that to see what my improvisational style looks like, so to speak.  And, I haven't done even this yet.

Re: Real Music Composition Software

Reply #6
That should make me feel guilty as hell for using Noteworthy Composer but never (really) having composed a single piece of musing. Lucky for me: guilt is one of my underdeveloped features.

I am not so sure of the musical outcome of a machine that does the hard work. It's a bit like church organs:  an electronic thing could never take its place... or could it? I have seen and heard Heijligers organs that are very, very impressive. Not quite the organs that you find in large churches and cathedrals, but still, of extremely good quality indeed. And "it could never be done, because..."

Well, the qualities that are asked of a "composing machine" are not within reach yet. But impossible? No.

Re: Real Music Composition Software

Reply #7
While I am also a skeptic on this subject, a search on 1/f and music turns up a number of interesting items, including:

http://www.pnas.org/cgi/reprint/88/8/3507

one of the more lucid discussions.

Re: Real Music Composition Software

Reply #8
I am not so sure of the musical outcome of a machine that does the hard work. It's a bit like church organs:  an electronic thing could never take its place... or could it? I have seen and heard Heijligers organs that are very, very impressive. Not quite the organs that you find in large churches and cathedrals, but still, of extremely good quality indeed. And "it could never be done, because..."
When our church was considering a new organ around 1994 or so, the organist then heard some high quality electronic instruments.  Her famous quote was "It sounds like a recording of a very good organ."  We eventually got a fine pipe organ from Casavant (sp?).
Since 1998

Re: Real Music Composition Software

Reply #9
George Gershwin was an expert at developing themes and motifs, not only within a single piece but across his entire range of works.
Yet Gershwin studied for 5 years under the guidance of Joseph Schillinger, the author of "The Schillinger System of Musical Composition", a method of musical composition based on mathematical principles, similar in scope to the composition system suggested by the original poster.
Of course Gershwin worked it all out on paper or at the keyboard, but do you think that he wouldn't have used a computer if one would have been available?
I'm no Gershwin - I don't have the inate ability to reject or accept mathematically generated variations on a basic theme - that's what made him a genius - but composition software and mathematical techniques are just another way of developing ideas.
Don't reject them out of hand.
Bill's right - It is a technique not an end in itself.
Schillinger himself wrote:
    "My system does not circumscribe the composer's freedom, but merely points out the methodological way to arrive at a decision. Any decision, which results in a harmonic relation, is fully acceptable. We are opposed only to vagueness and haphazard speculation."

Attached a piece developed by computer based Schillinger's principles for expanding on a theme. (Repost).

Re: Real Music Composition Software

Reply #10
And always remember that our greatly missed Fred Nachbar used fractal theme generators in writing some of his music.  And it wasn't chopped liver.

Re: Real Music Composition Software

Reply #11
"Band in a Box" has a "Melodist" function that probably does some things the original poster was seeking.
Here is an extract from the User's Guide: 

"Using the "Melodist" feature you can compose a new song in the style of your choice, complete with intro, chords, melody, arrangement etc.  All you have to do is pick from one of the many Melodists available and press the [OK] button"

I hadn't tried it out previously, but a quick trial a few minutes ago showed that there is a large range of different styles of "Melodist", nearly all "pop', "swing" and a wide range of jazz styles, nothing traditional or classical that I could see.  The results are quite pleasant, but maybe a bit in the "Muzak" class!

I must admit I haven't given it a fair work-out by any means.


Re: Real Music Composition Software

Reply #12
Let me be clear about this: there is nothing wrong in using a computer to generate inversions, retrogrades, sequences, etc. That's the grunt work, and any method that will lighten that load is helpful. NWC already does transpositions, checks for redundant accidentals, etc. - same idea.

Nor is there anything inherently wrong with mathematically-derived music. Computers have been used to do that, as I indicated earlier, since 1957; but it's been done much, much longer. Plato advocated it. In the 14th century, composers were heavily into mathematically-related techniques such as isorythm (the same rythmic pattern applied over and over to different melodic material, which might or might not be the same length as the rythmic pattern), crab canon (harmonizing a melody by using the same melody under it, but in reverse), and others. Guillaume de Machaut was the greatest master of this, but there were many others. In the first half of the 20th century, the total serialists came along, deriving everything mathematically - opposed by the stochastic composers, who did everything by throwing dice (or some equivalent chance method). Remnants of both of these schools persist today.

Some of the music produced by these methods is very beautiful, some is ugly but intellectually satisfying, and some is just plain ugly. The challenge is to sort these out.

That's where the composer's ear is necessary. And that, I think, is where computers will never be able to do the work. Computers operate by rules, and beauty doesn't. Simple as that.

Most of the music I write is based on a tone row, and one of the earliest stages of work on any piece of this type is to prepare a chart of all the forms of the row that will be used in the piece - original, retrograde, inversion, and retrograde inversion - in all possible transpositions. I do this (with a great deal of help from NWC). Since there is more than one way to arrange this chart, I always make at least two different forms of it, and sometimes three. I print these charts out, and while I am composing, whether at the computer or at the piano, I refer to the charts constantly. But the charts are not the music.

In that last sentence, I think, is where my argument with the original poster lies.

Cheers,

Bill

Re: Real Music Composition Software

Reply #13
Total agreement here.  Fred wrote that the fractal generator turned out dozens (hundreds?  I don't have his original comments to hand) of sequences from which he winnowed a few to work with.  He had a CD, Fractile, with thirteen of his fractal-based compositions.  I don't know if it is still available -- Google google --

Well, Dogstar music seems to be gone.  Part of the material on the old site is at:

http://www.dogstar.dantimax.dk/

but I don't see the music itself.  Lets try the wayback machine:

http://web.archive.org/web/20051210101030/http://www3.telus.net/dogstarmusic/index.htm

which seems to have the music files -- at least the NWC files, I didn't check the .mp3 files.

That is the last full copy of the dogstar site.  The only later one is a single page, the link for which fails.

I don't know if everything there is on the Scriptorium, so perhaps one needs to grab it now before it totally evaporates!