Processing Sound Using Photoshop
Sound designers, musicians, producers and engineers are all familiar with manipulating sound through the use of audio processing. Most examples of processing, like filters, reverbs, and delays produce a relatively predictable result. I'm interested in processing that has an unpredictable result. For example, the octave pedal that purrs like a robotic kitten when you feed it harmonic input, that I wrote about in Octave Pedal Rhodes. I have a similar interest in producing sound directly from visual media with tools like Michel Rouzic's Photosounder, which I used for Conversion of Graffiti into Sound.
Original Electric Piano Pattern (MP3)
One of the first things I tried to do with Photosounder was to capture the image it creates from imported audio in order to use an image editing application as an audio processor. This didn't work very well because much of the resolution was lost with the screen grab. Since then, Michel has added a feature to export full resolution spectral images of sound. This makes the concept of using a Photoshop filter as an audio processor possible.
Glowing Edges Electric Piano Pattern (MP3)
To test this concept I created a simple pattern with an electric piano patch and opened it in Photosounder. Without changing any settings I immediately saved the sound as a bitmap image. Next I opened the image in Photoshop and started experimenting with filters. Once I had some filtered images I loaded them back into Photosounder to see how they sounded. Gaussian blur and Liquefy created some unique effects, but my favorite of the bunch was Glowing Edges. This filter seems to transform the electric piano into a haunting choral passage. I'll be posting a similar version of this article on AudioCookbook.org as well. If anyone is class is interested in trying this out, let me know.
-John Keston
Comments
Wow, that is really interesting.. have you used this in real time at all? I would be really interested in seeing what would happen with live graffiti being feed into the program, if that would work at all. Also, can you use live instruments, such an feeding a guitar or keyboard into photosounder and then being able to manipulate it, instead of creating patches?
Posted by: kara | January 26, 2009 5:42 PM
Thanks John, good resource.
Makes me think of the old Metasynth days, remember that amazing thing?
http://www.uisoftware.com/MetaSynth/
Posted by: Ali Momeni | January 26, 2009 6:01 PM
Hey, Kara. Unfortunately at this time, there's no way to do this kind of processing in real-time. Photosounder does have visual parameters, like gamma and invert, etc. but they all take a few seconds to render the audio after processing the image; far too long for live performance.
Posted by: John Keston | January 26, 2009 7:31 PM
I haven't had a chance to check out MetaSynth, although the demo sounds are wicked.
Posted by: John Keston | January 26, 2009 7:37 PM
Kara : Photosounder as it currently is cannot do much in real-time or "live", however someone once scripted ARSS, the command-line predecessor of Photosounder, to periodically (every couple of seconds) feed an image from a webcam. They pointed the webcam to a blackboard and thusly created a strange beat in real-time by drawing on the blackboard and hearing the changes as it looped
http://www.lshift.net/blog/2008/07/25/listening-to-your-webcam#comment-6
Posted by: Michel Rouzic | January 27, 2009 12:42 AM