On Fri, 12 Nov 2010 16:25:56 -0600
grant centauri <[email protected]> wrote:

> okay, i think i have a better idea now of what you are going for.  i
> can't really offer a quick solution, but it will likely involve
> programming.  I think you're better off trying to start with a web
> cam or something like that as an input device.  I know that
> processing has libraries for accessing camera devices, I'm sure Pd
> probably does too... then its a matter of mapping graphical data to
> sound.
> 
> it seems like the software you're looking at, arss amd enscribe are
> mapping visual data to audio frequency data.  each one probably uses
> a different technique and method of mapping to get the sounds out of
> an image.  here's what enscribe does:
> 
> The scanlines of the input image are converted into frequency
> components and
> > then using an inverse Fast Fourier Transform, are converted into
> > sound. The left side of the image is the low frequency end, and the
> > right is the high end, up to just under the Nyquist limit if you
> > want it.
> >
> 
> there's a whole world out there of people doing this kind of stuff, I
> just found this:
> 
> I think any discussion of "transcoding" image (or other data) into
> sound has to take into account that such mapping is purely
> subjective. The artistry is in finding meaningful transpositions from
> one medium to another. Using "raw" data as sample values is certainly
> one way, and using an image as spectrographic values for resynthesis
> is certainly another. There are certainly enough softwares out there
> that do either trick, and those sounds are familiar to us from the
> many many artists who have used those softs already (to death in some
> cases). Dig a little deeper and decide for yourself the relationship
> of pixel to sample instead of relying on other people's ready-mades
> and you might be on to something new.
> 
> Basically, any image source used as a sound source is a matter of
> deciding how the data gets mapped.  A good place to start would
> probably be for you to design a mapping method for static images that
> makes sounds you like, and then explore various images and patterns
> that produce sound.  you could likely do this in Pd, but i'm not
> familiar enough with it to know how.
> 
> I know in the old days people used to 'animate' sound.  Films had the
> audio soundtrack actually printed on them as a waveform and a photo
> sensor would read that data to reproduce audio electronically.  Some
> animators started drawing their own patterns of light and dark in
> order to synthesize sound directly onto film.
> 
> I don't want to discourage your idea, i do think it is an interesting
> one... i'm imagining a scanner that you can set different images on
> to get different sounds, or control the speed and light sensitivity
> as you said... it could be pretty neat.  but i think it might be an
> uphill battle, and there's probably other ways you could put your
> graphical background to work making digital sound.
> 
> -grant
> 
> 
> 
> 2010/11/12 João Mário <[email protected]>
> 
> > On Thu, 11 Nov 2010 18:50:26 -0600
> > grant centauri <[email protected]> wrote:
> >
> > > i imagine what you really want to do is be able to somehow map
> > > data coming in from the scanner to some sort of musical or more
> > > interesting sound.  use scanner data as control data for some
> > > audio generator.
> > >
> > > i'm not sure how you'd manage that... i'm guessing it is possible
> > > with some level of programming knowledge.
> > >
> > > if that is the case let me know, then we can maybe attack this at
> > > a different angle.
> > >
> > > On Thu, Nov 11, 2010 at 6:23 PM, grant centauri
> > > <[email protected]> wrote:
> > >
> > > > cat simply dumps the output of one thing into the standard
> > > > output. if you type cat readme.txt you'll get the whole text of
> > > > the readme file pumped out into your terminal screen. if you do
> > > > cat mysong.mp3 you'll get a bunch of gobbledygook flooding your
> > > > terminal.
> > > >
> > > > /dev/dsp is the device name of your sound card.  so if you cat
> > > > the contents of a file or the output of another device you can
> > > > use the redirect command,
> > > > >, to send that output to /dev/dsp and you'll likely hear
> > > > >something.
> > > >
> > > > for example i just did this command:
> > > >
> > > > sudo cat audio/HumanAfterAll.wav > /dev/dsp
> > > >
> > > > and i'm now listening to a sort of ringing noisy mess.
> > > >
> > > > doing:
> > > >
> > > > sudo cat Desktop/americas_logo.png > /dev/dsp
> > > >
> > > > gave me like 15 seconds of whitish noise.
> > > >
> > > > if your scanner is a usb scanner, you can find out what device
> > > > it is by plugging it in and running this command in a terminal:
> > > >
> > > > dmesg | tail
> > > >
> > > > I just did it with a wireless mouse device and got this info:
> > > >
> > > > [17792.582043] usb 3-2: new low speed USB device using uhci_hcd
> > > > and address 2
> > > > [17792.746260] usb 3-2: configuration #1 chosen from 1 choice
> > > > [17792.781655] input: Wireless Mouse Wireless Mouse as
> > > > /devices/pci0000:00/0000:00:10.1/usb3/3-2/3-2:1.0/input/input6
> > > > [17792.781866] generic-usb 0003:05FE:0011.0002: input,hidraw1:
> > > > USB HID v1.10 Mouse [Wireless Mouse Wireless Mouse] on
> > > > usb-0000:00:10.1-2/input0
> > > >
> > > > i searched around my /dev directory and found this:
> > > >
> > > > /dev/bus/usb/003/002
> > > >
> > > > I take that as usb3-2
> > > >
> > > > your machine might be different, but the usb scanner should
> > > > come up as a device somewhere in /dev/bus/usb
> > > >
> > > > so then you could have a terminal open, run the cat command and
> > > > then control your scanner with whatever software you have on
> > > > your computer to scan images or something like that.
> > > >
> > > > its not very elegant, but you might get some noise out of it.
> > > >
> > > > 2010/11/11 João Mário <[email protected]>
> > > >
> > > > On Thu, 11 Nov 2010 23:45:13 +0000
> > > >> Rob Myers <[email protected]> wrote:
> > > >>
> > > >> > On 11/11/2010 11:29 PM, esphera wrote:
> > > >> > >
> > > >> > > Just wondering that you might understand me wrong the
> > > >> > > previous post. I want to know how can I make a scanner's
> > > >> > > input data, transformed into sound?
> > > >> > >
> > > >> > > I had some advices such as "hack the drivers" or "you can
> > > >> > > do it with puredata"
> > > >> > >
> > > >> > > but I would really like to know how?
> > > >> > >
> > > >> > > thanks very much
> > > >> > >
> > > >> > > p.s.: I already use arss and enscribe software. but I want
> > > >> > > hardware stuff.
> > > >> >
> > > >> > I don't think you'll be able to just do it with pure
> > > >> > hardware. Every scanner has a commmunication protocol, even
> > > >> > old-fashioned parallel port scanners. You need at least
> > > >> > something like an Arduino to send the commands the scanner
> > > >> > is listening for to start it scanning.
> > > >> >
> > > >> > Which isn't really related to pure:dyne...
> > > >> >
> > > >> > - Rob.
> > > >> >
> > > >> > ---
> > > >> > [email protected]
> > > >> > http://identi.ca/group/puredyne
> > > >> > irc://irc.goto10.org/puredyne
> > > >>
> > > >> It certainly isn't related to p:d but I thought that p:d
> > > >> community would fit in this kind of project... so I ask for
> > > >> opinion
> > > >>
> > > >> @grant centauri - what those commands do?
> > > >>
> > > >> ---
> > > >> [email protected]
> > > >> http://identi.ca/group/puredyne
> > > >> irc://irc.goto10.org/puredyne
> > > >>
> > > >
> > > >
> >
> > yes it's exactly that! :) like arss or enscribe (maybe this software
> > can be taken as study?..)
> > I don't know as I'm from image and type cult rather than
> > programming. I do have a strong and growing interest in electronics
> > and programming thanks to puredyne! and I would like very much to
> > see this work as a kind of instrument with some controls
> > over the speed of scan light or sensitivity... I want to go that
> > way if you read me. probably with pd and some midi controllers?
> >
> > ---
> > [email protected]
> > http://identi.ca/group/puredyne
> > irc://irc.goto10.org/puredyne
> >

hey man thanks for your time!

you really did touch the G point, the question of how data is mapped
has been revolving on my head. But then I think that the image as
spectogram is nice for what i'm looking for because I want to actually
see what kind of imagery makes what kind of sound (so that my
images production affects sound production and viceversa) but then urges
another question: I think the synthesis type is the thing that i'm
missing in understanding. arss uses both sine and noise synthesis.
enscribe i'm not sure. I got in touch with the creator and all he said
to me was :

"A lot more time and energy has gone into ARSS, I think they use
additive synthesis instead of iFFT, so the architectures are completely
different. enscribe was never intended for synthesis, just as a way to
hide images inside songs.

http://photosounder.com/examples.php also looks pretty cool, and  
there's a free demo."

What I said to him was that I like arss's the most for the kind of
clean sound results that it produced, but enscribe offers color mapping
related to panning and stuff, and that a fusion of both would be
perfect. But, I can't do it and i don't have the time to study about it.

So concluding this tought, I want to know what kind of synthesis is
more "pure" to what i'm looking for as a kind of synesthesia thing you
see?

Thanks a lot one more time for your time! I really expect more ideas :)

---
[email protected]
http://identi.ca/group/puredyne
irc://irc.goto10.org/puredyne

Reply via email to