Sunday, September 13, 2009
Basic Vocoding with Photoshop
YouTube via Photosounder
"http://photosounder.com
http://photosounder.com/blog/2009/09/...
This is how to perform basic vocoding with Photosounder and Photoshop. The carrier signal (a saw wave) and the modulator (HAL 9000's voice) are turned into images using Photosounder. Then in Photoshop, the voice is vertically blurred as to fuse the different harmonics together, then it's pasted on top of the image of the saw wave by multiplication. The result is then loaded into Photosounder in lossless mode with the sound of the carrier as a reference signal, giving the vocoded voice as a result."
7 comments:
Note: comments that insult people will be removed. Critique on gear is allowed. Do not ask if listings are still available. Click through auction links to check yourself. Posts and pics remain for historical purposes. To reduce spam, comments for posts older than one week are not displayed until approved (usually same day).
PREVIOUS PAGE
NEXT PAGE
HOME
© Matrixsynth - All posts are presented here for informative, historical and educative purposes as applicable within fair use.
MATRIXSYNTH is supported by affiliate links that use cookies to track clickthroughs and sales. See the privacy policy for details.
MATRIXSYNTH - EVERYTHING SYNTH
© Matrixsynth - All posts are presented here for informative, historical and educative purposes as applicable within fair use.
MATRIXSYNTH is supported by affiliate links that use cookies to track clickthroughs and sales. See the privacy policy for details.
MATRIXSYNTH - EVERYTHING SYNTH
Subscribe to:
Post Comments (Atom)
It's all the fun of a vocoder plus all the fun of manual labour and not being able to tweak anything in real time.
ReplyDeleteHaving a signal processing background, it just seems like unnecessary steps have been added without the spectacle of a full-on Rube Goldberg device.
I don't think it was supposed to be a substitute to a real vocoder, just a cool hack. Besides, it gives some idea of what a vocoder does.
ReplyDeleteVery interesting concept - I was reading today about the Cedar technology used to clean up the Beatles
ReplyDeletehttp://www.cedaraudio.com/news/ibs_lecture2009.html
and this looks like the same kind of idea if on a much more affordable scale Cedar plug ins are £1,5000 +). You probably need to play around for hours to get anything decent but looks promising.....
In the high quality mode I believe you could do Cedar retouching rather easily (he even has the tools in place to do this)...
ReplyDeleteAs for the vocoding illustration, I think what he's trying to do is show a technique that's familiar using his tool, the results that can be achived with the app are quite a bit more varied and having a few known 'tricks' like this in mind when exploring can provide a base of reference that can be...built around/modified/extended etc.
I remember when people said 'convolution is nice for weird sounds and all but I cant see it replacing my [multifx unit] for something that I can't even do in realtime.' Similar tales about "hey you're using that RONG it's supposed to be an automated BASS PLAYER"...etc.
scarr : Yep, not so practical for basic vocoding as demonstrated, but it also allows you to do what you can't do in real time, you can stretch around the modulator to fit your carrier! For example you can take bits of the voice and move or stretch them in time to fit a melody and its rhythm.
ReplyDeletePhilip - I've played with the Photosounder demo and this is nothing like what Cedar does. Photosounder is basically like Metasynth, in that it analyzes an image and plays back and audio file based on that image. The resolution isn't very high, so you could convert a sound to an image, do absolutely nothing to that image, then convert it back to audio and it will sound noticeably different - sort of cloudy and "ringy" do to the frequency spectrum being split up. Still, Photosounder can produce some cool effects. It's just not intended to "retouching".
ReplyDeletejabrockson : You're very mistaken, to do the Cedar-like stuff, press the lossless mode button ;). Also if you want more pixels in your image edit the file config.txt and increase the values for pixels_per_second and bands_per_octave to whatever you want (although I don't recommend that on sounds a few minutes long, it will create an image so huge no editor would be able to even display it ;-)).
ReplyDeleteAlso Cedar uses a linear frequency scale, which is alright for editing high frequency components but can be a PITA for lower frequencies, or seeing lower frequencies and higher frequencies at the same time.