Strange Ideas ;)

Let the developers know what you think of the software and what can be done to either improve the CFDG language or the Context Free program.

Moderators: MtnViewJohn, chris, mtnviewmark

Post Reply
shevegen
Posts: 57
Joined: Wed Jul 06, 2005 5:38 am

Strange Ideas ;)

Post by shevegen »

http://chriscoyne.com/gallery/view.php? ... 7718640000

Ok not sure if that link works. Chris made a cfdg that outputs song rules (Sheet Music).

Here the link to the img:
http://chriscoyne.com/gallery/uploads// ... 744216.png
Image

Ok, lately I saw this machine that automagically made music. Well, sure, it sound as if a 1 year old made noise but it was still funny to see. My question is whether there be tools that would attempt to correlate this image with real music - would be fun to hear it being played hehe. (Though I'd expect it to not sound very nice, it would be kinda unique though) :)

raimondious
Posts: 12
Joined: Thu Apr 13, 2006 3:45 pm

Post by raimondious »

Well, I wouldn't do it by inputting the image, but you could just write a CFG script (similar, but simpler than ContextFree) to generate a song, and then a player that took the output and played it...

SONG -> MOTIF SONG | STOP;
MOTIF -> PHRASE1 PHRASE2 | PHRASE2 PHRASE3;
PHRASE1 -> C F G [ha!] | C D E G;
PHRASE2 -> D A B C | B C F G;
PHRASE3 -> ....

And then whatever this randomly generated could be taken by some MIDI program that read off the notes and played them -- you'd also want to implement rhythm and such, but that's just putting complexity onto this grammar.

shevegen
Posts: 57
Joined: Wed Jul 06, 2005 5:38 am

Post by shevegen »

Interesting, Thanks.

I have another strange idea in a new thread by the way. =)

williamsharkey
Posts: 1
Joined: Mon Jan 07, 2008 2:20 pm

Post by williamsharkey »

I'm interested in generating music on the raw level. Without human- or midi- interpreters.

I have two ideas going on:

1. Generate a 1pixel by 30,000 pixel bitmap image, and convert the image to a sound wave with matlab. The Pixel darkness would be mapped to speaker cone displacement. The horizontal would be mapped to time.

2. Generate a square or circle image in context free. Then write a software "record player" that once again maps speaker cone displacement to pixel darkness. The sampler (needle if you will) would take a circle path around the image. Just like a real record player.


I am new to CF, so I will need to create some basic functions. Like generating a sine wave using idea(1).

User avatar
mtnviewmark
Site Admin
Posts: 81
Joined: Wed May 04, 2005 12:46 pm
Location: Mountain View, CA
Contact:

Post by mtnviewmark »

You should look into the works of Herbert Brün, specifically his computer music works from 1972 called Dust, More Dust, etc.

He worked with computer generated music by building up whole compositions from the tiniest of samples. I think he would have loved a CFDG sort of way of doing this.....

I had the pleasure of meeting him in the early 1980s at a concert of his work in Boston. I was immediately entranced with his work because most of the computer music being produced at the time was "squeaky clean" with all these perfect tones and reverbs and timbre control, whereas Brün's Dust was full of distortion and grit. It was wonderful!

See also Granular Synthesis.
I'm the "m" in "mtree.cfdg"

Post Reply