Freesound Forums

Dear Freesounders,

Time has been in very short supply for me these last few weeks, so it has taken me ages to actually put this together...
But finally here it is:

The Sound Experiment (Dare 26 spin-off)

In Dare 26 we have been using some unusual sound generation techniques. Some of the sounds can be quite harsh and unpleasant, and a few people mentioned to me that those sounds made them feel 'uncomfortable'.
With this ound here, probably being one of the worst offenders:
http://www.freesound.org/people/luckylittleraven/sounds/223075/

This then lead me to some thought and some discussions: why do these sounds feel so wierd and unpleasant?
First of all, what makes a sound 'unpleasant'? And then, why do the sounds generated by this method sound particularly unpleasant?

I have a few theories of why that is. Now I need your help: I need a few pairs of ears to listen to several variants of the sound which I have created and rated them as "more unpleasant", "less unpleasant" or "same" as the original sound.
The sounds are in this pack:
http://www.freesound.org/people/AlienXXX/packs/14358/
Instructions of how to rate the sound are in the description of each sound.
I know there are quite a few sounds in this pack and they are not exactly pleasant to listen to. So I do not expect someone to go through all the sounds in the pack. But pick a few at random and leave your comment (as per instructions on the sound description)

What else? What next?
If you want to know a bit more about the experiment, read the text below.
The pack mentioned above contains a set of sounds, all derived from the original sound. Different processing techniques were used. These techniques were chosen according to sound science and my own hipothesys about what should make the sound more or less unpleasant.
I have detailed records of how each sample was created. But, in order not to influence the listener's oppinion, those are (currently) not mentioned in each sound description. Each sound was also given a non-descriptive name (leter + number) which gives no indication about how the sound was processed. - This was done using a randomization table, so I can then trace each sound back to how it was created.

The experiment will be open for a period of time. - How long exactly depends on interest and participation.
When the experiment closes:
I will post here to notity you.
The sound descriptions will be updated to indicate how each sound was created.
I will analyze the results, compare with my initial hipothesys and post here also.

Hope you enjoy being part of the experiment!

I just went through them all. I find it interesting that the ones noone else commented on were the ones I found most boring. Not unpleasant, just boring.

kb7clx does have a point: Some of the sounds are indeed "plain" and not very interesting. And since many of the sounds are similar, listening to a lot of them in sequence must be quite boring.
I apologize for that and I thank you all for your patience and effort in participating.

I am afraid that, as with many experiments, the experimentation itself is repetitive and boring and the exciting bit will be the analysis of the results.

toiletrolltube wrote:

Firstly. I don't think there is such a thing as a "boring" sound, all sounds have potential for creative purposes.

Now it's my turn to have a listen to the sounds.

toiletrolltube wrote:
God, the amount of times I've told myself to stay away from freesound when drinking.

Interesting experiment, can't wait to hear the analysis. It seems so far that try as you did none of the sounds were as hideous as the original ! I'd like to take the credit for that but really it was my computer, using the techniques you guys taught me
I'm gonna be having nightmares about that feedback . . . I'd like to apologise for the original sound was recorded way too loud (sorry) which definitely adds to the unpleasantness. I listened to the sounds through headphones and speakers and although the headphones pick up more detail I was more physically uncomfortable with the sounds coming out of the speakers. Something about these kinds of sounds - vibration/frequency/energy is physically jarring. But as this is a science experiment I will keep my somewhat hippy opinions to myself (for now)

Hi everyone.. how is life?
I just listened to a few samples and noticed something weird...
On a few samples there is this very high pitch sound... seems not everyone is able to hear this..
My girlfriend is listening with me and she didnt hear anything like this!!!
The frequency is very very high!
I know that when your hearing is becoming less you start to lose the higher frequencies first so this isnt an unknown phenomenon.
Only... the sound is so NASTY! beyond any limit.. that I decided to stop listening for tonight!
I just cant take anymore... it raises the hair in my neck... gosebumps all over!!
Just leaves me with two questions... do you hear this? And does it sound as nasty to you as it sounds to me?

@escortmarius

another very important thing is the system you are listening from: headphones, hi-fi stereo, professional monitors?
I guess you are not using headphones, as you mention your girlfriend is listening with you... But, if you are there is a lot of different types: in-ear, closed.end, open-end...

Another thing to try is to download one or two of the files and listen to the downloaded file, rather than the compressed version straight from the Freesound page and check if the high freq thing is still there.

Is it just 1 freq (kind of high pitched sine wave) or is it a mix of frequencies? (probably hard to tell, if it is really high freq).
If you want to find the frequency download one of the files. LOWER THE VOLUME. Stick an EQ in front of it with a narrow band pass filter. For safety, stick a limiter in front of the EQ, aimed at -6dB or even -10dB. - then sweep the EQ frequendy until you find that annoying sound.

Note that some EQs either stop bellow 20kHz or have a squashed frequency display at high frequencies. If your EQ does not go right up to high frequencies, try NyQuist EQ, from here:
http://magnus.smartelectronix.com/

Let us know what happens...

I am listening on my hifi system..
I upgraded my gear a little.. I have a Tascam US-428 interface now thats connected to my hifi amp.
This is one of the samples:
http://www.freesound.org/people/AlienXXX/sounds/234027/

Tonight I will download a few samples and see what I can find using an EQ...

Yes I hear it, the seemingly quieter second half of the sample is much worse than the first 'noisy' bit. The high pitched ringing in your ears sound, the awful clicking and static feeling, the droning noise and some weird silent noise that I can only describe as uncomfortable pressure.
it will be interesting to hear your investigation as I have no idea how to sweep a frequency but I might have to try it out one day

I'm really enjoying trying these processing techniques and pondering 'what is a sound?' I'm not really sure on the use for raw data sounds either, but they're so unique even if they are generated. I made a few 3d studio scenes in bryce so I'd have some bmp files to import as data and experiment with. The original sample in this experiment 'menacing machine' was a rocky river!

You can trust your ears and gears
RTA's like Voxengo SPAN clearly show a peak around 14.7-14.9 KHz that actually increases in the second part of the mentioned sound, making it clip from second ~17 to ~17.5

@toiletrolltube
Yes, I did get it. You made me smile with that comment. I thought it was funny.

@copyc4at
I must say this was a bit of an unnexpected development. IT is making the discussion much more interesting than I expected. I would imagine, however, that a peak at 14.7-14.9kHz would be eard by most people. I guess I need to check.

@luckylittleraven and toiletrolltube
As for the use of data sounds... Some of the short beeps and blips can be used on their own, as they are and can be sequenced (i.e., played in a pattern). Think the kind of sound on early computer games "blip blip brrrrr blip blip blooop...
Some more extreme variants of noise music can use everything. Including large amounts of unprocessed sounds from databending. - everything is possible!...

But most sounds, especially longer sounds, tend to sound harsh and fatigating on the ears. Not musical at allThere are many techniques that can be used to run those sounds into non-pitched (usually meaning percussion, but also noise, drones and effects) sounds and even into pitched sounds (i.e., "sounds that hod a note"). - We can explore some of these techniques in detail (now we have plenty of raw material sounds to work with!)
So, toiletrolltube is right: they are great for processing. Good raw material.

copyc4t wrote:
You can trust your ears and gears
RTA's like Voxengo SPAN clearly show a peak around 14.7-14.9 KHz that actually increases in the second part of the mentioned sound, making it clip from second ~17 to ~17.5

funny enough, we have on today's sound-of-the-day one example of what I meant about short databending sounds programmed into a little sequence:
http://www.freesound.org/people/rap2h/sounds/151537/
I am not sure if these sounds came from databending. But they are the kind of sounds that could have been produced that way.

While you are at it. Press the "similar sounds" button. (you can find it on the right of the page. Look for the download button... further down... past the file technical details... THERE - Press it!)
So... the Freesound algorythm things this sound is similar to several circuit-bent speak-n-spell sounds. Interesting, since circuit-bending these devices is, in a way, corrupting the data that is the sound being played from the memory chip... When circuit bending this kind of device you often end up playing parts of the ROM that do not contain "sound"... So you can be playing DATA.
One of these circuit bent sounds took me to this pack. If you have a few minutes listen. Amazing what can be done with a circuit-bent toy and some effects...
http://www.freesound.org/people/IvoryBunker/packs/14384/

Finally, here is a musical use of databent samples (thanks and credits to Trebblofang/MetaDronos)
http://soundcloud.com/metadronos/cynordrulenun
Probably not something you would dance to, but a very listenable piece that could be the soundtrack to some heavy movie scenes. A bit creepy, heh!?

I think I just de-railed my own thread
... should have posted this on one of the initial Dare 26 threads.
Ah well...

Lets get it on track again....

To find out what was going on AlienXXX suggested to run the samples thrue an EQ to see what was going on.
The sound to me was more or less the same on all the samples that had this high pitch sound, so I used the following sample:

http://www.freesound.org/people/AlienXXX/sounds/234027/

I found out that there was a very high peak at 14,8kHz around 14sec. and again at 20sec.
After that I maxed out my EQ at 14,8kHz. setting everything to maximize the peak that is so very very very annoying!!!
And noticed that when I zoned in on that frequency the sound became less annoing!
I made this sample in the process:

http://www.freesound.org/people/escortmarius/sounds/234505/

Curious about what you all think/hear..

Seems like we have established you have the most sensitive hearing of us all.

I do hear the high frequency sound you mention. Also in my sample that you posted above.
But for me it is low intensity and not annoying at all. It is actually one of the least offending aspects of the sample - The starting sound on that original sample is just horrible for me. Lasts until about 6s,, then an additional low frequency aspect comes in and makes the sound even worse (if that was possible!). This lasts until about 11s and then fades leaving only a 'clicking' (I find this very annoying).
From 13s onwards (the area where you hear the high freq sound I hear only a high freq hiss and chirping. Low vol. Not annoying at all. Like the hiss you hear on the soundtrack of an old movie or an old record. From about 16s the clicking starts again (there is a lot of very low freq contents in this sample as well). I find the clicking very unpleasant on headphones.
From about 19s, the low frequency 'rumble' starts to appear again. I hear the high pitch chirping mixed in with that, but again, I do not find it annoying at all.
From 19-20s onwards is actually the most bearable part of this sample, if you ask me.

On your filtered sample, I have put the headphone volume on max just now.
Just a mild, low fol chirping. Like crickets and cicadas on a summer night. No problem for me at all to listen to this. Your filtered sample is 100 times nicer to my ears than the sample you started with.

By the way, do you find the sound of crickets/cicadas very annoying? (I mean the actual insects 'playing' live, not recordings of them.)

That's amazing how you pinpointed it like that! You can clearly hear it start then it dies down for a few seconds then goes again for the rest of the sample. It's like an electrical ringing/whining, that you get on old things like tvs and consoles
It looks loud in the wave form, but it's not as 'loud' as the awful noise at the beginning of the original. Not stereo-typically loud but maddening! I haven't played it through speakers, that would be pretty mean because my cat's sitting next to me . . . but it tickled in the headphones, not that loud was too loud. I also hear the clicking, and you seem to feel it too, like fluttering which is quite disconcerting. I wonder if the unbearable pressure or silent noise is the low frequency rumble that AlienXXX mentioned.
I'm really not good at analysing frequencies and I only know the basic physics of sound, I just know I felt very threatened by this sound. hence 'menacing machine'

AlienXXX wrote:
On your filtered sample, I have put the headphone volume on max just now.
Just a mild, low fol chirping. Like crickets and cicadas on a summer night. No problem for me at all to listen to this. Your filtered sample is 100 times nicer to my ears than the sample you started with.

By the way, do you find the sound of crickets/cicadas very annoying? (I mean the actual insects 'playing' live, not recordings of them.)

Be carefull with the volume!!! I maxed out the EQ and volume to get the most effect possible...
Dont know, but it could be bad for your ears!! Dont want anyone to get hearing problems....

I like the sound of crickets... but thats also because I know that they are crickets..
The noise of machines is nothing strange to me... I work at a metal company every day so I am used to noises like grinding drilling pressing welding etc etc..
Maybe thats also why I hear that specific frequency so well... it stands out from what I am used to.