View Single Post
Old 11-07-2013, 07:28 AM   #1216
InfiniteIce
Questionist
 
Join Date: Nov 2013
Posts: 8
Here's how I figured out/got this.

The sound/song playing on the page is tinp.wav in the source. It took me a few days to wonder why he would use a .wav, the filesize is so huge, and this file does not need to be lossless...i was going to see what the encoder info was. Maybe it was just named wav because of compatibility with his webhost or something. So I downloaded it. Why would he server a 33MB wav file..?

Open it up in a spectral analyzer to create a spectrogram. Any DAW like Audition or Ableton will do this if you have one, but if you don't, there are free ones but they aren't as nice to look at, change colors, or zoom as much. Audacity is a free DAW.

If you JUST want something free and fast to show the spectrals and never plan to do any kind of audio getting, get "Spek" or "Spectro" (two different programs) from www.spek.cc & spectro.enpts.com respectively. I have used both, they are safe and work fine.

You'll see a bunch of 'sound waves', more or less (distribution of frequency of a sound clip.). The color of them might be awful depending on the program you chose. Sometimes it can be changed (high-contrast helps see details better.) Zoom in on the waveform..

You are presented with a shadow of a man in the middle. On the left and right, to me look like pairs of hands reaching out toward the man. The hands on the right side look like they are in jail or behind bars.

There are a few examples of this in commercial CDs. The most famous is probably from Aphex Twin. Nine Inch Nails has done one as well (easy to find on google. Or just look up "famous specrograms/graphs.

Does anybody recognize anything about the image? I will probably try it in a few different colors to see if anything else pops out at me.


Things I have tried: Splitting the file into two based on stereo positions (So, basically get a waveform for the left speaker, and one for the right.) Playing just one speaker reveals nothing new. I've tried filtering down the noise to decipher the words. I've tried taking out the 'picture' via editing but nothing useful. Tried playing with AudioPaint for a little bit.

Tangential but probably very related things in figuring out its meaning:
imagesynth. just google it.
Audio Steganography, which means "..the art and science of encoding hidden messages in sadio formats in such a way that no one, apart from the sender and intended recipient, suspects it.."

Stuff that probably isn't relevant but I'll post just in case. tinp.wav was made in ProTools and last encoded on 2012-10-08 at 01:00:09 AM. It was encoded at 48kHz for some reason. It definitely didn't need to be. Might just be the default protools option, I'm not sure.

http://vanilla47.com/Link%20Collecti...esMethods.html

Last edited by InfiniteIce; 11-07-2013 at 07:46 AM.
InfiniteIce is offline   Reply With Quote