hi,
I'd like to be able to adjust the settings of bonk~ to the (spectral) features of the sound of clapping in your hands but the documentation in bonk-help about the learning mode is a bit incomplete I think. For example if I use the debounce message bonk~ stops outputting any values... The article from ICMC about fiddle and bonk is very interesting in general but very short about the specific operation mode of bonk.
i use bonk~ like this:
- i send it a [learn xx( (where xx is the number of times i present a
sample of one sound - like clapping)
- then i present the sound xx times, then i present another sound xx
times, etc until i presented all the sounds (might be knocking, clapping, i even tried with saying letters like p,b,s)
- then i give it a [learn 0( message and test it - if im happy with the
result i write it to a file..
i dont know if this helped you, though :) probably you should change any variables during learning..
gnd/
I've tried clapping into mics and cant get consistent results -- the burst of air goes in all different directions and I could never get a consistent sound into the mic at all. But if you can solve _that_ problem, I think gnd's idea should work OK.
cheers Miller
On Mon, Oct 12, 2009 at 10:43:32AM -0000, gnd@itchybit.org wrote:
hi,
I'd like to be able to adjust the settings of bonk~ to the (spectral) features of the sound of clapping in your hands but the documentation in bonk-help about the learning mode is a bit incomplete I think. For example if I use the debounce message bonk~ stops outputting any values... The article from ICMC about fiddle and bonk is very interesting in general but very short about the specific operation mode of bonk.
i use bonk~ like this:
- i send it a [learn xx( (where xx is the number of times i present a
sample of one sound - like clapping)
- then i present the sound xx times, then i present another sound xx
times, etc until i presented all the sounds (might be knocking, clapping, i even tried with saying letters like p,b,s)
- then i give it a [learn 0( message and test it - if im happy with the
result i write it to a file..
i dont know if this helped you, though :) probably you should change any variables during learning..
gnd/
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Try a large diaphragm dynamic mic like the Shure SM7 or old Beyer RE20 which take fast transient high dB sounds well. Put a pop filter in front of the mic and either compress the signal for a smoother envelope or gate it just to get the attack of the clap.
On Mon, Oct 12, 2009 at 7:57 PM, Miller Puckette mpuckett@imusic1.ucsd.eduwrote:
I've tried clapping into mics and cant get consistent results -- the burst of air goes in all different directions and I could never get a consistent sound into the mic at all. But if you can solve _that_ problem, I think gnd's idea should work OK.
cheers Miller
On Mon, 12 Oct 2009, Miller Puckette wrote:
I've tried clapping into mics and cant get consistent results -- the burst of air goes in all different directions and I could never get a consistent sound into the mic at all.
"burst of air" sounds like a very low frequency impulse... if you can't rely on that part of the sound to be the same, you have to remove it, so that you are left with the part of the sound that is reliable, so that [bonk~] reacts reliably.
if the very low freqs are too much for the mic... because the lower freqs have inane amplitudes for their amount of energy... then what physical device constitutes an acoustic highpass filter that can remove the distorsion ? something that e.g. can remove much of the effect of wind blowing into a mic when trying to record outdoors... i don't know that stuff.
...
i just looked up "pop filter" when seeing that word in cgc's reply, and it seems like it's a highpass filter, though it doesn't use those words in the description I read, but I guess it from what they say about clipping and aspirated plosives.
http://en.wikipedia.org/wiki/Pop_filter
but this article is weird, because it mentions the hiss but doesn't say what a pop filter does about hiss...
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard, Montréal, Québec. téléphone: +1.514.383.3801
I've tried clapping into mics and cant get consistent results -- the burst of air goes in all different directions and I could never get a consistent sound into the mic at all.
You'll get that if you try to close mic claps. Assuming a studio environment, moving the mic back mitigates that problem. I suggest one meter away as a starting point.
Cheers,
Marvin Humphrey
Mathieu Bouchard wrote:
On Mon, 12 Oct 2009, Miller Puckette wrote:
I've tried clapping into mics and cant get consistent results -- the burst of air goes in all different directions and I could never get a consistent sound into the mic at all.
...
i just looked up "pop filter" when seeing that word in cgc's reply, and it seems like it's a highpass filter, though it doesn't use those words in the description I read, but I guess it from what they say about clipping and aspirated plosives.
http://en.wikipedia.org/wiki/Pop_filter
but this article is weird, because it mentions the hiss but doesn't say what a pop filter does about hiss...
It is not so much a question of highpass vs. lowpass as it is a directional filtering - the pop filter attenuates the most direct path of the sound, and if I am not mistaken they are designed to attenuate more with a higher energy burst of sound (this attenuation achieved via air turbulence), in order to filter more strongly on the loudest sibilants (high pitched) and plosives (low pitched) while affecting the quieter sounds as little as possible.
This turbulence theory may be wrong, I looked for confirmation or denial online but my google skills are failing me.
Hi
the pop filter does not do anything against the hiss, it is just trying to take away the plosives, like low frequency high power airstream, which couses the membran of the microphone to react in not wished/calculated ways.
so with the turbolence theory you are quiet on the right way.
der.brandt
Mathieu Bouchard wrote:
On Mon, 12 Oct 2009, Miller Puckette wrote:
I've tried clapping into mics and cant get consistent results -- the burst of air goes in all different directions and I could never get a consistent sound into the mic at all.
...
i just looked up "pop filter" when seeing that word in cgc's reply, and it seems like it's a highpass filter, though it doesn't use those words in the description I read, but I guess it from what they say about clipping and aspirated plosives.
http://en.wikipedia.org/wiki/Pop_filter
but this article is weird, because it mentions the hiss but doesn't say what a pop filter does about hiss...
It is not so much a question of highpass vs. lowpass as it is a directional filtering - the pop filter attenuates the most direct path of the sound, and if I am not mistaken they are designed to attenuate more with a higher energy burst of sound (this attenuation achieved via air turbulence), in order to filter more strongly on the loudest sibilants (high pitched) and plosives (low pitched) while affecting the quieter sounds as little as possible.
This turbulence theory may be wrong, I looked for confirmation or denial online but my google skills are failing me.
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
On Tue, 13 Oct 2009, Justin Glenn Smith wrote:
Mathieu Bouchard wrote:
i just looked up "pop filter" when seeing that word in cgc's reply, and it seems like it's a highpass filter, though it doesn't use those words in the description I read, but I guess it from what they say about clipping and aspirated plosives.
It is not so much a question of highpass vs. lowpass as it is a directional filtering - the pop filter attenuates the most direct path of the sound
Attenuation is normally not frequency-independent, unless it is designed to be so. So pretty much anything has an highpass and/or lowpass aspect to it unless you compensate it or you're just lucky...
I can't imagine studio mics doing things that favour the bounced-off waves at the detriment of the direct sound...
and if I am not mistaken they are designed to attenuate more with a higher energy burst of sound (this attenuation achieved via air turbulence),
Uh, wouldn't air turbulence would be a consequence (side-effect) of the filtering, not the cause of it?
This turbulence theory may be wrong, I looked for confirmation or denial online but my google skills are failing me.
In the case of sibilants, there is a turbulence going on inside of the mouth, but once it comes out of the mouth, it becomes just noise. Pretty much any large-scale natural [noise~]-like sound has to be generated by some kind of turbulence. Turbulence is normally a generator of noise, isn't it?
_ _ __ ___ _____ ________ _____________ _____________________ ... | Mathieu Bouchard, Montréal, Québec. téléphone: +1.514.383.3801
-
brandt＠subnet.at -
chris clepper -
gnd＠itchybit.org -
Justin Glenn Smith -
Marvin Humphrey -
Mathieu Bouchard -
Mike Moser-Booth -
Miller Puckette