Freesound Forums

Also - I redid my original single sine wave noise sample, but with phase matching. And it sounds identical to the original. It seems that all the digital glitches are actually masked by the rapidly changing frequencies.

Thanks luckylittleraven for that holysheet comment LOLX. After reading your comment I downloaded audiopaint to just check that what is it.... And really it is awesomatic Experience.....

I've also uploaded this one which may be useful to the OP. It sounds much closer to white noise than I expected. But I'm happy to prove myself wrong(!)
http://www.freesound.org/people/Speedenza/sounds/241464/

@AlienXXX - thanks for your comments on both sounds. On the temporal sound experiment http://www.freesound.org/people/Speedenza/sounds/241457/ I'll see if I can make something along the lines you suggest - I'm sure you're right about the frequency effect.

Yes, letting the spacing and distribution of your actual oscillator frequencies have some randomness should get you much closer to white noise. Nice demo. But if you don't have some randomness in there, somehow, somewhere, then it ain't right.

Using just audiopaint, I created one-second segments built with increasing counts of oscillators spread exponentially, and the artifacts of the mathematical relationships between the frequencies became more and more pronounced as the the number of oscillators increased. It sounded less and less like an out-of-tune organ rather quickly, but I went up to 9999 oscillators, and it still sounded crappy. I was using 40Hz-18KHz as the spread. And then, finally, I realized that the exponential spacing would be approximating pink noise, not white noise, and so I changed to linear spacing on the 9999 -sized image to get it more "white". Indeed, it was a bit like white, but with horrible artifacts, sounding almost like some weapon out of Star Trek (TOS). I'm not sure it's worth putting something like that up on freesound since it really has no value, but it's an easy experiment to do if you want. On my 32-bit OS, however, audio-paint would crash (out of memory) if reading a 20x2200 pixel image, so I had to go the other way (2200x20) and then rotate it 90 degrees once it was in memory (and that worked). I only went up to 9999x20. (the 20 was arbitrary).

zimbot wrote:
I'm not sure it's worth putting something like that up on freesound since it really has no value, but it's an easy experiment to do if you want.

Even for those who do have access to the software, reading through the forum thread and having immediate access to the sound examples being described is a lot more interesting and convenient than having to go away and try to replicate what they reading about.

Speedenza wrote:
I've also uploaded this one which may be useful to the OP. It sounds much closer to white noise than I expected. But I'm happy to prove myself wrong(!)
http://www.freesound.org/people/Speedenza/sounds/241464/

AlienXXX of course you're right, but I'm lazy. They're uploaded now.
whiteCombined.wav (forgot to change the file name).
white9999linear.wav