stAllio! on tumblr |
i make audio collage, glitch art, and so on. i've been databending since the '90s, which either means i'm old-school or simply that i'm old. |
yes, it is very possible to transcode audio or other data into image files. (in fact, i thought i had written about this before, but i can’t turn it up if so.) depending on your tech skills, there are any number of ways to do this; i’ll describe how to do it in photoshop.
under the photoshop file menu, select File > Open As. this lets you trick photoshop into thinking you’re loading a different sort of data than you are. in the Open As dialog box, find the Open As drop-down at the bottom and select Photoshop Raw. then pick a file to open and click OK.
next you’ll see the Photoshop Raw Options dialog, which has some important choices. photoshop will try to “guess” how big the image is based on file size, and often comes up with a suggestion immediately. generally, this first guess will have only one color channel (in other words, grayscale or “black and white”). if grayscale sounds cool, then great: click OK and you’re good to go.
if you want “full color” then you likely want RGB, which has 3 color channels, so you’ll want to enter a 3 in the Channel box, and make a selection for either interleave or non-interleave (they look different).
this next part can get tedious. you need to come up with size settings that will make sense to photoshop, so you’ll be using the Guess button. delete the numbers in the Width and Height boxes, if any. enter a 0 in the Header, and then click the Guess button. if you’re lucky, numbers will appear in the Height and Width boxes, in which case you can click OK. if no numbers appear, then enter a 1 in the Header box and click Guess. if that doesn’t work, enter a 2… keep incrementing the number in the Header box until the Guess button generates numbers for you. when it does, click OK to see your transcoded data.
of course, you can take more control over it… if you wanted a specific height or width, you could enter that while leaving the other box blank. or you could even pick a specific image size and try to use the Guess button to generate the correct number of header bytes. this might speed things up or slow them down, but the process is mostly the same.
like i said, there are lots of other ways you could do this, too. for example, evan meaney has a new project called null sets which uses scripting to convert texts to jpegs.
PS: since you asked about my sound collage work, you can find my latest tracks at soundcloud or my website, and most of my releases (as well as those from my band, animals within animals) are available for free from badtaste4life.com
soundcloud sent me a takedown notice because they think one of my datasound tracks — where every sound in the song is taken from sonified data — contains a sample of some house record nobody’s ever heard of. this is so glitch i can hardly believe it really happened.
dubstep producer joker. original photo by alex de mora.
glitches commissioned by front magazine, issue 163
erotic science
Go see Mario.
He knows erotic science
When you are young you make love naturally