Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.os.linux.misc > #12130 > unrolled thread

White Noise Software For Linux

Back to article view | Back to comp.os.linux.misc

Contents

  White Noise Software For Linux beforewisdom@gmail.com - 2014-09-19 16:41 -0700
    Re: White Noise Software For Linux William Unruh <unruh@invalid.ca> - 2014-09-20 06:39 +0000
      Re: White Noise Software For Linux The Natural Philosopher <tnp@invalid.invalid> - 2014-09-20 09:32 +0100
        Re: White Noise Software For Linux floyd@apaflo.com (Floyd L. Davidson) - 2014-09-20 02:44 -0800
          Re: White Noise Software For Linux The Natural Philosopher <tnp@invalid.invalid> - 2014-09-20 11:53 +0100
            Re: White Noise Software For Linux floyd@apaflo.com (Floyd L. Davidson) - 2014-09-20 04:25 -0800
              Re: White Noise Software For Linux Michael Black <et472@ncf.ca> - 2014-09-20 11:52 -0400
          Re: White Noise Software For Linux William Unruh <unruh@invalid.ca> - 2014-09-22 18:31 +0000
            Re: White Noise Software For Linux floyd@apaflo.com (Floyd L. Davidson) - 2014-09-22 18:59 -0800
              Re: White Noise Software For Linux Michael Black <et472@ncf.ca> - 2014-09-22 23:30 -0400
                Re: White Noise Software For Linux floyd@apaflo.com (Floyd L. Davidson) - 2014-09-23 02:51 -0800
                  Re: White Noise Software For Linux William Unruh <unruh@invalid.ca> - 2014-09-23 23:49 +0000
    Re: White Noise Software For Linux asdf <asdf@nospam.com> - 2014-09-21 09:55 +0000
      Re: White Noise Software For Linux Rich <rich@example.invalid> - 2014-09-21 12:41 +0000
        Re: White Noise Software For Linux JohnF <john@please.see.sig.for.email.com> - 2014-10-02 23:56 +0000
          Re: White Noise Software For Linux Rich <rich@example.invalid> - 2014-10-03 01:41 +0000
            Re: White Noise Software For Linux JohnF <john@please.see.sig.for.email.com> - 2014-10-03 02:13 +0000
              Re: White Noise Software For Linux Rich <rich@example.invalid> - 2014-10-03 02:55 +0000
                Re: White Noise Software For Linux JohnF <john@please.see.sig.for.email.com> - 2014-10-03 05:38 +0000
                  Re: White Noise Software For Linux JohnF <john@please.see.sig.for.email.com> - 2014-10-03 11:50 +0000
                    Re: White Noise Software For Linux Loki Harfagr <l0k1@thedarkdesign.free.fr.INVALID> - 2014-10-07 13:04 +0000
          Re: White Noise Software For Linux William Unruh <unruh@invalid.ca> - 2014-10-03 14:38 +0000
            Re: White Noise Software For Linux JohnF <john@please.see.sig.for.email.com> - 2014-10-08 03:18 +0000

Page 1 of 2  [1] 2  Next page →

#12130 — White Noise Software For Linux

I'm running Linux Mint 17.

Can anyone recommend any white noise software that is controlled by a GUI?

Thanks much

Steve

[toc] | [next] | [standalone]

#12131

On 2014-09-19, beforewisdom@gmail.com <beforewisdom@gmail.com> wrote:
> I'm running Linux Mint 17.
>
> Can anyone recommend any white noise software that is controlled by a GUI?

Wha is White noise sofware tha is controlled by a gui? White noise has
only one controllable parameter and that is ampliude, and tha is
conrollable by the volume control.
I think you need o be a bit less elliptical.
>
> Thanks much
>
> Steve

[toc] | [prev] | [next] | [standalone]

#12132

On 20/09/14 07:39, William Unruh wrote:
> On 2014-09-19, beforewisdom@gmail.com <beforewisdom@gmail.com> wrote:
>> I'm running Linux Mint 17.
>>
>> Can anyone recommend any white noise software that is controlled by a GUI?
>
> Wha is White noise sofware tha is controlled by a gui? White noise has
> only one controllable parameter and that is ampliude, and tha is
> conrollable by the volume control.
> I think you need o be a bit less elliptical.
Thnks for sayung that. saved me the trouble

>>
>> Thanks much
>>
>> Steve


-- 
Everything you read in newspapers is absolutely true, except for the 
rare story of which you happen to have first-hand knowledge. – Erwin Knoll

[toc] | [prev] | [next] | [standalone]

#12133

The Natural Philosopher <tnp@invalid.invalid> wrote:
>On 20/09/14 07:39, William Unruh wrote:
>> On 2014-09-19, beforewisdom@gmail.com <beforewisdom@gmail.com> wrote:
>>> I'm running Linux Mint 17.
>>>
>>> Can anyone recommend any white noise software that is controlled by a GUI?
>>
>> Wha is White noise sofware tha is controlled by a gui? White noise has
>> only one controllable parameter and that is ampliude, and tha is
>> conrollable by the volume control.
>> I think you need o be a bit less elliptical.
>Thnks for sayung that. saved me the trouble

White noise is not restricted to just one controllable parameter.

Distribution is controllable, hence it might be Gaussian or Poisson, or
something else.

It is also always bandwidth limited, which is controllable too.

-- 
Floyd L. Davidson                         http://www.apaflo.com/
Ukpeagvik (Barrow, Alaska)                      floyd@apaflo.com

[toc] | [prev] | [next] | [standalone]

#12134

On 20/09/14 11:44, Floyd L. Davidson wrote:
> The Natural Philosopher <tnp@invalid.invalid> wrote:
>> On 20/09/14 07:39, William Unruh wrote:
>>> On 2014-09-19, beforewisdom@gmail.com <beforewisdom@gmail.com> wrote:
>>>> I'm running Linux Mint 17.
>>>>
>>>> Can anyone recommend any white noise software that is controlled by a GUI?
>>>
>>> Wha is White noise sofware tha is controlled by a gui? White noise has
>>> only one controllable parameter and that is ampliude, and tha is
>>> conrollable by the volume control.
>>> I think you need o be a bit less elliptical.
>> Thnks for sayung that. saved me the trouble
>
> White noise is not restricted to just one controllable parameter.
>
> Distribution is controllable, hence it might be Gaussian or Poisson, or
> something else.
>
then its not white, its colored

> It is also always bandwidth limited, which is controllable too.
>

then its not white.


-- 
Everything you read in newspapers is absolutely true, except for the 
rare story of which you happen to have first-hand knowledge. – Erwin Knoll

[toc] | [prev] | [next] | [standalone]

#12135

The Natural Philosopher <tnp@invalid.invalid> wrote:
>On 20/09/14 11:44, Floyd L. Davidson wrote:
>> The Natural Philosopher <tnp@invalid.invalid> wrote:
>>> On 20/09/14 07:39, William Unruh wrote:
>>>> On 2014-09-19, beforewisdom@gmail.com <beforewisdom@gmail.com> wrote:
>>>>> I'm running Linux Mint 17.
>>>>>
>>>>> Can anyone recommend any white noise software that is controlled by a GUI?
>>>>
>>>> Wha is White noise sofware tha is controlled by a gui? White noise has
>>>> only one controllable parameter and that is ampliude, and tha is
>>>> conrollable by the volume control.
>>>> I think you need o be a bit less elliptical.
>>> Thnks for sayung that. saved me the trouble
>>
>> White noise is not restricted to just one controllable parameter.
>>
>> Distribution is controllable, hence it might be Gaussian or Poisson, or
>> something else.
>>
>then its not white, its colored
>
>> It is also always bandwidth limited, which is controllable too.
>>
>
>then its not white.

The definition of "white noise" doesn't agree with those statements.

-- 
Floyd L. Davidson                         http://www.apaflo.com/
Ukpeagvik (Barrow, Alaska)                      floyd@apaflo.com

[toc] | [prev] | [next] | [standalone]

#12136

On Sat, 20 Sep 2014, Floyd L. Davidson wrote:

> The Natural Philosopher <tnp@invalid.invalid> wrote:
>> On 20/09/14 11:44, Floyd L. Davidson wrote:
>>> The Natural Philosopher <tnp@invalid.invalid> wrote:
>>>> On 20/09/14 07:39, William Unruh wrote:
>>>>> On 2014-09-19, beforewisdom@gmail.com <beforewisdom@gmail.com> wrote:
>>>>>> I'm running Linux Mint 17.
>>>>>>
>>>>>> Can anyone recommend any white noise software that is controlled by a GUI?
>>>>>
>>>>> Wha is White noise sofware tha is controlled by a gui? White noise has
>>>>> only one controllable parameter and that is ampliude, and tha is
>>>>> conrollable by the volume control.
>>>>> I think you need o be a bit less elliptical.
>>>> Thnks for sayung that. saved me the trouble
>>>
>>> White noise is not restricted to just one controllable parameter.
>>>
>>> Distribution is controllable, hence it might be Gaussian or Poisson, or
>>> something else.
>>>
>> then its not white, its colored
>>
>>> It is also always bandwidth limited, which is controllable too.
>>>
>>
>> then its not white.
>
> The definition of "white noise" doesn't agree with those statements.
>
You can use a long shift register and suitable feedback to make a 
pseudo-random number generator, and run that fast enough and you get 
"white noise".  There was an IC made for that purpose at one point (and 
the same technique was used in things like the Commodore SID and the 
AY-4-8910 sound generator), and it would give you "white noise" the caveat 
being it probably was limited output the higher in frequency you got, but 
likely fine for the intended purpose of audio.

You can reverse bias a diode and generate white noise.  That works at 
audio, and it works at microwave, though better to have a better diode if 
you want higher frequencies.

But, this isn't a "let's limit bandwidth" so much as "this is good enough 
for the purposes at these frequencies".  White noise by any definition 
I've seen is broadband.

But, if you want "pink noise" for testing audio equipment, you start with 
a white noise generator (like that pseudo-random shift register setup) and 
a filter that falls off a certain amount per octave.  But that's no longer 
white noise.

   Michael

[toc] | [prev] | [next] | [standalone]

#12145

On 2014-09-20, Floyd L. Davidson <floyd@apaflo.com> wrote:
> The Natural Philosopher <tnp@invalid.invalid> wrote:
>>On 20/09/14 07:39, William Unruh wrote:
>>> On 2014-09-19, beforewisdom@gmail.com <beforewisdom@gmail.com> wrote:
>>>> I'm running Linux Mint 17.
>>>>
>>>> Can anyone recommend any white noise software that is controlled by a GUI?
>>>
>>> Wha is White noise sofware tha is controlled by a gui? White noise has
>>> only one controllable parameter and that is ampliude, and tha is
>>> conrollable by the volume control.
>>> I think you need o be a bit less elliptical.
>>Thnks for sayung that. saved me the trouble
>
> White noise is not restricted to just one controllable parameter.
>
> Distribution is controllable, hence it might be Gaussian or Poisson, or
> something else.

Then it is not white noise. It may be noise, but white noise has a very
particular definition
>
> It is also always bandwidth limited, which is controllable too.

Again, not white noise. White noise is noise such that at each instant
of time, the noise is uncorrelated with that at the next instant of
time. And at that instant of time it is gaussian. 

Anyway the OP has still not told us what he wants. 

>

[toc] | [prev] | [next] | [standalone]

#12147

William Unruh <unruh@invalid.ca> wrote:
>On 2014-09-20, Floyd L. Davidson <floyd@apaflo.com> wrote:
>> The Natural Philosopher <tnp@invalid.invalid> wrote:
>>>On 20/09/14 07:39, William Unruh wrote:
>>>> On 2014-09-19, beforewisdom@gmail.com <beforewisdom@gmail.com> wrote:
>>>>> I'm running Linux Mint 17.
>>>>>
>>>>> Can anyone recommend any white noise software that is controlled by a GUI?
>>>>
>>>> Wha is White noise sofware tha is controlled by a gui? White noise has
>>>> only one controllable parameter and that is ampliude, and tha is
>>>> conrollable by the volume control.
>>>> I think you need o be a bit less elliptical.
>>>Thnks for sayung that. saved me the trouble
>>
>> White noise is not restricted to just one controllable parameter.
>>
>> Distribution is controllable, hence it might be Gaussian or Poisson, or
>> something else.
>
>Then it is not white noise. It may be noise, but white noise has a very
>particular definition

Look it up though, and you'll find "Gaussian White Noise"... and other
distributions too.

http://en.wikipedia.org/wiki/White_noise#Statistical_properties

>> It is also always bandwidth limited, which is controllable too.
>
>Again, not white noise.

It is impossible to have a signal that is not bandwidth limited.
Real life just doesn't work that way.

>White noise is noise such that at each instant
>of time, the noise is uncorrelated with that at the next instant of
>time. And at that instant of time it is gaussian.

Gaussian does not imply white noise nor does white noise imply Gaussian.

>Anyway the OP has still not told us what he wants.

-- 
Floyd L. Davidson                         http://www.apaflo.com/
Ukpeagvik (Barrow, Alaska)                      floyd@apaflo.com

[toc] | [prev] | [next] | [standalone]

#12148

On Mon, 22 Sep 2014, Floyd L. Davidson wrote:

> William Unruh <unruh@invalid.ca> wrote:
>> On 2014-09-20, Floyd L. Davidson <floyd@apaflo.com> wrote:
>>> The Natural Philosopher <tnp@invalid.invalid> wrote:
>>>> On 20/09/14 07:39, William Unruh wrote:
>>>>> On 2014-09-19, beforewisdom@gmail.com <beforewisdom@gmail.com> wrote:
>>>>>> I'm running Linux Mint 17.
>>>>>>
>>>>>> Can anyone recommend any white noise software that is controlled by a GUI?
>>>>>
>>>>> Wha is White noise sofware tha is controlled by a gui? White noise has
>>>>> only one controllable parameter and that is ampliude, and tha is
>>>>> conrollable by the volume control.
>>>>> I think you need o be a bit less elliptical.
>>>> Thnks for sayung that. saved me the trouble
>>>
>>> White noise is not restricted to just one controllable parameter.
>>>
>>> Distribution is controllable, hence it might be Gaussian or Poisson, or
>>> something else.
>>
>> Then it is not white noise. It may be noise, but white noise has a very
>> particular definition
>
> Look it up though, and you'll find "Gaussian White Noise"... and other
> distributions too.
>
> http://en.wikipedia.org/wiki/White_noise#Statistical_properties
>
>>> It is also always bandwidth limited, which is controllable too.
>>
>> Again, not white noise.
>
> It is impossible to have a signal that is not bandwidth limited.
> Real life just doesn't work that way.
>
That's true, but it is based on the method used to generate the noise.
And you can get "reasonably close" so long as you use a method appropriate 
for the need.

Like I said, a pseudo-random number generator using a shift register with 
proper feedback clocked at a high enough frequency can generate "white 
noise" good enough for audio purposes.  And then you filter it properly if
you want "pink noise" which is more suited for some testing purposes.

But once you filter, it's not white, it's something else.

The limits of generation are a different thing.

   Michael

[toc] | [prev] | [next] | [standalone]

#12150

Michael Black <et472@ncf.ca> wrote:
>On Mon, 22 Sep 2014, Floyd L. Davidson wrote:
>
>> William Unruh <unruh@invalid.ca> wrote:
>>> On 2014-09-20, Floyd L. Davidson <floyd@apaflo.com> wrote:
>>>> The Natural Philosopher <tnp@invalid.invalid> wrote:
>>>>> On 20/09/14 07:39, William Unruh wrote:
>>>>>> On 2014-09-19, beforewisdom@gmail.com <beforewisdom@gmail.com> wrote:
>>>>>>> I'm running Linux Mint 17.
>>>>>>>
>>>>>>> Can anyone recommend any white noise software that is controlled by a GUI?
>>>>>>
>>>>>> Wha is White noise sofware tha is controlled by a gui? White noise has
>>>>>> only one controllable parameter and that is ampliude, and tha is
>>>>>> conrollable by the volume control.
>>>>>> I think you need o be a bit less elliptical.
>>>>> Thnks for sayung that. saved me the trouble
>>>>
>>>> White noise is not restricted to just one controllable parameter.
>>>>
>>>> Distribution is controllable, hence it might be Gaussian or Poisson, or
>>>> something else.
>>>
>>> Then it is not white noise. It may be noise, but white noise has a very
>>> particular definition
>>
>> Look it up though, and you'll find "Gaussian White Noise"... and other
>> distributions too.
>>
>> http://en.wikipedia.org/wiki/White_noise#Statistical_properties
>>
>>>> It is also always bandwidth limited, which is controllable too.
>>>
>>> Again, not white noise.
>>
>> It is impossible to have a signal that is not bandwidth limited.
>> Real life just doesn't work that way.
>>
>That's true, but it is based on the method used to generate the noise.
>And you can get "reasonably close" so long as you use a
>method appropriate for the need.

Any "method appropriate" means it is in fact bandwidth limited.

The bandwidth is simply controlled at something larger than the
perceived need.

>Like I said, a pseudo-random number generator using a
>shift register with proper feedback clocked at a high
>enough frequency can generate "white noise" good enough
>for audio purposes.

Is by definition bandwidth limited.

Goot enouogh for audio means that it is extremely limited!

>And then you filter it properly if
>you want "pink noise" which is more suited for some testing purposes.
>
>But once you filter, it's not white, it's something else.

That depends on the filter.  In your example the initial
filter doesn't stop it from being white noise, but certainly
filtering to produce pink noise does.

>The limits of generation are a different thing.

Not at all.  It's just another filter.

-- 
Floyd L. Davidson                         http://www.apaflo.com/
Ukpeagvik (Barrow, Alaska)                      floyd@apaflo.com

[toc] | [prev] | [next] | [standalone]

#12155

On 2014-09-23, Floyd L. Davidson <floyd@apaflo.com> wrote:
>

<Much nattering on about the definition of white noise removed>
We still have no idea of what the OP wanted. Until they tell us that
this discussion is not going to go anywhere. 

[toc] | [prev] | [next] | [standalone]

#12138

beforewisdom@gmail.com wrote:

> I'm running Linux Mint 17.
>
> Can anyone recommend any white noise software that is controlled by a GUI?

dd if=/dev/urandom of=/dev/youroutputaudiodevice behind a xmessage
block?:)

Seriously, I seem to recall a plugin or some option for Audacity to
generate white noise.

Take also a look here.
http://kokkinizita.linuxaudio.org/linuxaudio/

[toc] | [prev] | [next] | [standalone]

#12139

asdf <asdf@nospam.com> wrote:
> beforewisdom@gmail.com wrote:

> > I'm running Linux Mint 17.
> >
> > Can anyone recommend any white noise software that is controlled by a GUI?

> dd if=/dev/urandom of=/dev/youroutputaudiodevice behind a xmessage
> block?:)

> Seriously, I seem to recall a plugin or some option for Audacity to
> generate white noise.

> Take also a look here.
> http://kokkinizita.linuxaudio.org/linuxaudio/

sox has a large number of "noise" (both white and colored) options.

Here is one that someone posted somewhere on Usenet in the not too
distant past to generate brown noise:

play -q -c 2 -n synth brownnoise band -n 1600 1500 tremolo .1 30

[toc] | [prev] | [next] | [standalone]

#12195

Rich <rich@example.invalid> wrote:
> beforewisdom@gmail.com wrote:
> 
>> Can anyone recommend any white noise software
> 
> sox has a large number of "noise" (both white and colored) options.
> Here is one that someone posted somewhere on Usenet in the not too
> distant past to generate brown noise:
>   play -q -c 2 -n synth brownnoise band -n 1600 1500 tremolo .1 30

Thanks, that works great. And I see from man sox (or man play)
that output can be saved to a file. And I suppose it's got to be
pretty obvious, but I'm not seeing (or getting it to work
by messing around) exactly how to do that with your above example.
That is, suppose I want to generate, say, a 1GB file containing
the above, something functionally like the second line below...
  play -q -c 2 -n synth brownnoise band -n 1600 1500 tremolo .1 30 \
       --size 1GB --format mp3 -o brownnoise.mp3
How would I correctly do that? Thanks again,
-- 
John Forkosh  ( mailto:  j@f.com  where j=john and f=forkosh )

[toc] | [prev] | [next] | [standalone]

#12197

JohnF <john@please.see.sig.for.email.com> wrote:
> Rich <rich@example.invalid> wrote:
> > beforewisdom@gmail.com wrote:
> > 
> >> Can anyone recommend any white noise software
> > 
> > sox has a large number of "noise" (both white and colored) options.
> > Here is one that someone posted somewhere on Usenet in the not too
> > distant past to generate brown noise:
> >   play -q -c 2 -n synth brownnoise band -n 1600 1500 tremolo .1 30

> Thanks, that works great. And I see from man sox (or man play)
> that output can be saved to a file. And I suppose it's got to be
> pretty obvious, but I'm not seeing (or getting it to work
> by messing around) exactly how to do that with your above example.
> That is, suppose I want to generate, say, a 1GB file containing
> the above, something functionally like the second line below...
>   play -q -c 2 -n synth brownnoise band -n 1600 1500 tremolo .1 30 \
>        --size 1GB --format mp3 -o brownnoise.mp3
> How would I correctly do that? Thanks again,

After many false starts (sox has a very convoluted CLI interface),, I
found the "hint" in the docs to make it work:

  sox -q -c 2 -n out.wav synth brownnoise band -n 1600 1500 tremolo .1 30

Produces brownnoise in out.wav.

What it does not do is "stop" at a certian length.  You could watch the
file in another window, and kill the sox command when the file reaches
the length you want.  If you really want to directly output a specific
length, then this will work (sox will complain that the wav header
length value will be incorrect, but the file seems to play anyway):

  sox -q -c 2 -n -t wav - synth brownnoise band -n 1600 1500 tremolo .1 30 | dd of=out.wav bs=1024 count=1000

bs = block size
count = number of blocks

So total size is bs*count, adjust to the size you wish to obtain.

[toc] | [prev] | [next] | [standalone]

#12198

Rich <rich@example.invalid> wrote:
> JohnF <john@please.see.sig.for.email.com> wrote:
>> Rich <rich@example.invalid> wrote:
>> > beforewisdom@gmail.com wrote:
>> > 
>> >> Can anyone recommend any white noise software
>> > 
>> > sox has a large number of "noise" (both white and colored) options.
>> > Here is one that someone posted somewhere on Usenet in the not too
>> > distant past to generate brown noise:
>> >   play -q -c 2 -n synth brownnoise band -n 1600 1500 tremolo .1 30
> 
>> Thanks, that works great. And I see from man sox (or man play)
>> that output can be saved to a file. And I suppose it's got to be
>> pretty obvious, but I'm not seeing (or getting it to work
>> by messing around) exactly how to do that with your above example.
>> That is, suppose I want to generate, say, a 1GB file containing
>> the above, something functionally like the second line below...
>>   play -q -c 2 -n synth brownnoise band -n 1600 1500 tremolo .1 30 \
>>        --size 1GB --format mp3 -o brownnoise.mp3
>> How would I correctly do that? Thanks again,
> 
> After many false starts (sox has a very convoluted CLI interface),
> I found the "hint" in the docs to make it work:
>   sox -q -c 2 -n out.wav synth brownnoise band -n 1600 1500 tremolo .1 30
> Produces brownnoise in out.wav.

Thanks so much, Rich. That works perfectly. I thought I'd looked at
that -n in man sox, and it said null device/file. Seems weird that
it works, but maybe I shouldn't be asking too many questions:).

> What it does not do is "stop" at a certain length. You could watch the
> file in another window, and kill the sox command when the file reaches
> the length you want. If you really want to directly output a specific
> length, then this will work (sox will complain that the wav header
> length value will be incorrect, but the file seems to play anyway):
>    sox -q -c 2 -n -t wav - synth brownnoise band -n 1600 1500 tremolo .1 30
>    | dd of=out.wav bs=1024 count=1000
> bs = block size
> count = number of blocks
> So total size is bs*count, adjust to the size you wish to obtain.

Thanks, again. Kind of "my bad" about that size thing, which I just
threw in without thinking much. Very sorry if it made you work more.
Ctrl-C is pretty much okay for me, or maybe --seconds for duration
of play would've been a more sensible -switch.
   By the way, I let it run for maybe ~10secs (on an i7-3770K), and
out.wav contained ~448MB. I was curious to see if sox ended abruptly
or terminated the file properly after Ctrl-C, but haven't listened
all the way to the end yet:). Thanks,
-- 
John Forkosh  ( mailto:  j@f.com  where j=john and f=forkosh )

[toc] | [prev] | [next] | [standalone]

#12200

JohnF <john@please.see.sig.for.email.com> wrote:
> Rich <rich@example.invalid> wrote:
> > JohnF <john@please.see.sig.for.email.com> wrote:
> >> Rich <rich@example.invalid> wrote:
> >> > beforewisdom@gmail.com wrote:
> >> > 
> >> >> Can anyone recommend any white noise software
> >> > 
> >> > sox has a large number of "noise" (both white and colored) options.
> >> > Here is one that someone posted somewhere on Usenet in the not too
> >> > distant past to generate brown noise:
> >> >   play -q -c 2 -n synth brownnoise band -n 1600 1500 tremolo .1 30
> > 
> >> Thanks, that works great. And I see from man sox (or man play)
> >> that output can be saved to a file. And I suppose it's got to be
> >> pretty obvious, but I'm not seeing (or getting it to work
> >> by messing around) exactly how to do that with your above example.
> >> That is, suppose I want to generate, say, a 1GB file containing
> >> the above, something functionally like the second line below...
> >>   play -q -c 2 -n synth brownnoise band -n 1600 1500 tremolo .1 30 \
> >>        --size 1GB --format mp3 -o brownnoise.mp3
> >> How would I correctly do that? Thanks again,
> > 
> > After many false starts (sox has a very convoluted CLI interface),
> > I found the "hint" in the docs to make it work:
> >   sox -q -c 2 -n out.wav synth brownnoise band -n 1600 1500 tremolo .1 30
> > Produces brownnoise in out.wav.

> Thanks so much, Rich. That works perfectly. I thought I'd looked at
> that -n in man sox, and it said null device/file. Seems weird that
> it works, but maybe I shouldn't be asking too many questions:).

Sox has an odd ordering of parameters:

  sox [global stuff] [input file] [output file] [effects]

> > What it does not do is "stop" at a certain length. You could watch the
> > file in another window, and kill the sox command when the file reaches
> > the length you want. If you really want to directly output a specific
> > length, then this will work (sox will complain that the wav header
> > length value will be incorrect, but the file seems to play anyway):
> >    sox -q -c 2 -n -t wav - synth brownnoise band -n 1600 1500 tremolo .1 30
> >    | dd of=out.wav bs=1024 count=1000
> > bs = block size
> > count = number of blocks
> > So total size is bs*count, adjust to the size you wish to obtain.

> Thanks, again. Kind of "my bad" about that size thing, which I just
> threw in without thinking much. Very sorry if it made you work more.

Nah, how to "get X amount" was the easy part, once I figured out how to
get it to "write to a file".  I'd already encountered the hyphen for
stdout (which is quite common) in the manpage.  So once I knew where
the output file needed to be, clipping the result was easy.

> Ctrl-C is pretty much okay for me, or maybe --seconds for duration
> of play would've been a more sensible -switch.
>    By the way, I let it run for maybe ~10secs (on an i7-3770K), and
> out.wav contained ~448MB. I was curious to see if sox ended abruptly
> or terminated the file properly after Ctrl-C, but haven't listened
> all the way to the end yet:). Thanks,

^C likely results in a final sample that is clipped.  The "dd" method
wil clip the final sample also (unless you compute number of samples +
header byte lenght and set a specifc dd size first, or get very lucky
in your size choice).

[toc] | [prev] | [next] | [standalone]

#12201

Rich <rich@example.invalid> wrote:
> JohnF <john@please.see.sig.for.email.com> wrote:
>> Rich <rich@example.invalid> wrote:
>> > JohnF <john@please.see.sig.for.email.com> wrote:
>> >> Rich <rich@example.invalid> wrote:
>> >> > beforewisdom@gmail.com wrote:
>> >> > 
>> >> >> Can anyone recommend any white noise software
>> >> > 
>> >> > sox has a large number of "noise" (both white and colored) options.
>> >> > Here is one that someone posted somewhere on Usenet in the not too
>> >> > distant past to generate brown noise:
>> >> >   play -q -c 2 -n synth brownnoise band -n 1600 1500 tremolo .1 30
>> > 
>> >> Thanks, that works great. And I see from man sox (or man play)
>> >> that output can be saved to a file. And I suppose it's got to be
>> >> pretty obvious, but I'm not seeing (or getting it to work
>> >> by messing around) exactly how to do that with your above example.
>> >> That is, suppose I want to generate, say, a 1GB file containing
>> >> the above, something functionally like the second line below...
>> >>   play -q -c 2 -n synth brownnoise band -n 1600 1500 tremolo .1 30 \
>> >>        --size 1GB --format mp3 -o brownnoise.mp3
>> >> How would I correctly do that? Thanks again,
>> > 
>> > After many false starts (sox has a very convoluted CLI interface),
>> > I found the "hint" in the docs to make it work:
>> >  sox -q -c 2 -n out.wav synth brownnoise band -n 1600 1500 tremolo .1 30
>> > Produces brownnoise in out.wav.
> 
>> Thanks so much, Rich. That works perfectly. I thought I'd looked at
>> that -n in man sox, and it said null device/file. Seems weird that
>> it works, but maybe I shouldn't be asking too many questions:).
> 
> Sox has an odd ordering of parameters:
>  sox [global stuff] [input file] [output file] [effects]

Thanks, Rich. That makes a lot more sense now.
I didn't figure it out from browsing the long sox man page.

>> > What it does not do is "stop" at a certain length. You could watch the
>> > file in another window, and kill the sox command when the file reaches
>> > the length you want. If you really want to directly output a specific
>> > length, then this will work (sox will complain that the wav header
>> > length value will be incorrect, but the file seems to play anyway):
>> >  sox -q -c 2 -n -t wav - synth brownnoise band -n 1600 1500 tremolo .1 30
>> >  | dd of=out.wav bs=1024 count=1000
>> > bs = block size
>> > count = number of blocks
>> > So total size is bs*count, adjust to the size you wish to obtain.
> 
>> Thanks, again. Kind of "my bad" about that size thing, which I just
>> threw in without thinking much. Very sorry if it made you work more.
> 
> Nah, how to "get X amount" was the easy part, once I figured out how to
> get it to "write to a file".  I'd already encountered the hyphen for
> stdout (which is quite common) in the manpage.  So once I knew where
> the output file needed to be, clipping the result was easy.
> 
>> Ctrl-C is pretty much okay for me, or maybe --seconds for duration
>> of play would've been a more sensible -switch.
>>    By the way, I let it run for maybe ~10secs (on an i7-3770K), and
>> out.wav contained ~448MB. I was curious to see if sox ended abruptly
>> or terminated the file properly after Ctrl-C, but haven't listened
>> all the way to the end yet:). Thanks,
> 
> ^C likely results in a final sample that is clipped.  The "dd" method
> wil clip the final sample also (unless you compute number of samples +
> header byte length and set a specifc dd size first, or get very lucky
> in your size choice).

I used lame to convert to mp3, and that didn't emit any complaints,
so I guess the .wav's okay, or okay enough. After messing around
a bit, what I ended up using was
   sox -q -c 1 -n white.wav synth noise band -n 900 1400
   lame white.wav white.mp3
and I suppose sox's output could be piped to lame's input,
avoiding the intermediate .wav file, but from discussion above
it might be wiser to avoid trying to figure that out:).
By the way, this is for a whitenoise generator, using...
 bhphotovideo.com/c/product/679046-REG/Pyle_Pro_PCAU22_PCAU22_Mini_40_Watt.html
 bhphotovideo.com/c/product/597576-REG/Pyle_Pro_PCB4BK_PCB4_200W_4_Mini.html
(url's just illustrate what I'm using; not (dis)recommending this seller,
and there's a later PCAU25A). The PCAU22 (and 25A) have usb slots,
and just continue repeating whatever's on the stick.
Right now, I have an old whitenoise generator (whose "native" volume
is too low) plugged into amp's input, but the usb stick will be
a lot neater. Thanks again,
-- 
John Forkosh  ( mailto:  j@f.com  where j=john and f=forkosh )

[toc] | [prev] | [next] | [standalone]

#12204

JohnF <john@please.see.sig.for.email.com> wrote:
> <<snip>> After messing around a bit, what I ended up using was
>   sox -q -c 1 -n white.wav synth noise band -n 900 1400
>   lame white.wav white.mp3

After messing around a little more, I like
    sox -q -c 1 -n white.wav synth noise band -n 24 800 echo .1 5 1 .99
a little better. Not sure there actually is such a thing as
better sounding white noise, but seems that way to me.
-- 
John Forkosh  ( mailto:  j@f.com  where j=john and f=forkosh )

[toc] | [prev] | [next] | [standalone]

Page 1 of 2  [1] 2  Next page →

Back to top | Article view | comp.os.linux.misc

csiph-web