A friend of mine is currently writing her diploma and for the presentation there are 6 speakers.
Each one of them should have a different, recognizable, but artistacly stimulating, visualization of his/her voice.
I tried to make a patch, but it is not looking very good, stylewise.
My friend, is willing to pay, how much I dont know, I guess not too much, but fair.
If anyone is interested in a little job for the portfolio please contact me.Or just answer in the thread.
I attached my patch as far as I got. but the presentation is in 2 weeks and some tweaking needs to be done.
If vvvv cannot handle 6 simultanus audio streams then there would be the possibilty that someone activates each style simply by clicking in the patch. Would be great if someone could help me out here.
A friend of mine is currently writing her diploma and for the presentation there are 6 speakers.
Each one of them should have a different, recognizable, but artistacly stimulating, visualization of his/her voice.
Just not sure, you mention 6 speakers, then visualization of voice, so do you mean 6 displays, or having 6 speakers (like a 5.1 setup where each channel has the sound transformed for example).
I tried to make a patch, but it is not looking very good, stylewise.
I attached my patch as far as I got. but the presentation is in 2 weeks and some tweaking needs to be done.
Had a look (quick) at the patch. From what I saw on the FFT output you do a sum then divide by 250 (basically you do a mean).
You can Simply use Mean (Spectral) instead.
Also fft data is between 0 and 1 (if sound is normalized), so it you connect your average to a spread count you will have no display (unless you map range the value, but would still be better to have a fixed spread count and change the size of the circle depending on fft for example.
If vvvv cannot handle 6 simultanus audio streams then there would be the possibilty that someone activates each style simply by clicking in the patch. Would be great if someone could help me out here.
Please refer to first section.
But if it’s display only you can easily use one FFT then output it to each of the 6 visuals.
well with"speakers" I was thinking of orators. People that give a speech :-)
There is just one Display/Beamer. And each man or woman that is talking should have a different visual style.
As people tend to talk contemporarily it would be great to hook up 6 audio streams into vvvv. I just dont know whether that is possible or not. Short example: person a gets a circle…
person b a line; person c a dot so if person a is talking the circle starts changing its size, colour, shading whatever. If then person be interupts a line appears or gets faded in …
If start argueing you can see both the circle and the line.
I will include your suggested changes in the patch tomorrow.
If any of you is interested in “styling” the 6 different “voices”
please let me know. I am afraid of loosing to much time…
do you have 6 microphones?
do you have a soundcard with 6 inputs?
i dont think so.
you could buy 6 cheapo 2,-- € microphones and connect them e.g. to an arduinoboard. use this only for sensing who speaks.
get input from main audio system and analize with fft,scope, etc. activate this audio influence on individual patches with your cheapo solution from above.
Actually, yes I do have 6 mics and currently just a Edirol Fa-66
(only 4 audio inputs) but a friend of mine is having the bigger brother with more inputs.
But that is not the question (although I feel a bit insulted, just why in the world shouldnt I have 6 mics and soundcard?? What is so odd about that ?)
The question was wether one of you out there has any experience with 6 simultanius audio streams in vvvv ?
And how to set them up any hints, tricks whatever…
Again the work will be paid …
I keep on trying.
Maybe someone can “lend” me some visual patches I can fit in?
I dont wanna rob anything. It is just for the diploma of a good friend.
I’m not sure how your card would work for driver, but normally you would need to set the input to Mono (Monaural in the AudioIn node).
In the driver Pin, for some card you can choose the input number.
Then you should be able to get fft from that.
Easy way to disable an “orator” is to compute the average of the fft, and disable his visual if it’s under a certain value (as there’s some background noise it will never be zero).
Can’t help you much more on that as I don’t have a multichannel sound card.
