Nulling files

classic Classic list List threaded Threaded
13 messages Options
Reply | Threaded
Open this post in threaded view
|

Nulling files

Mew, Peter
Hi
I have done the following using sox
Split a 44/16 file called originalfile into a number of pieces.
Recombined them and reversed the polarity with
Sox inputfiles outputfile vol -1
Then mixed the output file with the original with
Sox -m originalfile outputfile combinedfile
And then got stats for the combinedfile with
Sox combinedfile -n stats
And I'm looking at pk level dB in the stats
Although the combinedfile sounds perfectly nulled, stats reports pk level dB as -7
Other files I've tried usually null to around -80
I would expect nulls to -95 or so
Any ideas why this file should sound ok but not get reported correctly in stats.
Is it me reading the wrong parameter from stats, or using the wrong sox commands to accomplish this Nulling function

Thanks
-pm

Sent from my iPad

________________________________

“**Confidentiality**
The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.

We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Sox-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/sox-users
Reply | Threaded
Open this post in threaded view
|

Re: Nulling files

Mew, Peter
This What I get from stats in the terminal

             Overall     Left      Right
DC offset  -0.000000 -0.000000 -0.000000
Min level  -0.440674 -0.400543 -0.440674
Max level   0.440918  0.416229  0.440918
Pk lev dB      -7.11     -7.61     -7.11
RMS lev dB    -30.81    -30.20    -31.52
RMS Pk dB     -15.61    -15.61    -16.20
RMS Tr dB     -96.87    -96.84    -96.87
Crest factor       -     13.47     16.60
Flat factor     0.00      0.00      0.00
Pk count           2         2         2
Bit-depth      15/16     15/16     15/16
Num samples    29.2M
Length s     661.240
Scale max   1.000000
Window s       0.050

-pm
________________________________________
From: Mew, Pete [[hidden email]]
Sent: Wednesday, May 21, 2014 7:16 PM
To: [hidden email]
Subject: [SoX-users] Nulling files

Hi
I have done the following using sox
Split a 44/16 file called originalfile into a number of pieces.
Recombined them and reversed the polarity with
Sox inputfiles outputfile vol -1
Then mixed the output file with the original with
Sox -m originalfile outputfile combinedfile
And then got stats for the combinedfile with
Sox combinedfile -n stats
And I'm looking at pk level dB in the stats
Although the combinedfile sounds perfectly nulled, stats reports pk level dB as -7
Other files I've tried usually null to around -80
I would expect nulls to -95 or so
Any ideas why this file should sound ok but not get reported correctly in stats.
Is it me reading the wrong parameter from stats, or using the wrong sox commands to accomplish this Nulling function

Thanks
-pm

Sent from my iPad

________________________________

“**Confidentiality**
The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.

We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Sox-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/sox-users

________________________________

“**Confidentiality**
The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.

We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.

------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Sox-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/sox-users
Reply | Threaded
Open this post in threaded view
|

Re: Nulling files

Jan Stary
On May 21 19:16:35, [hidden email] wrote:
> I have done the following using sox
> Split a 44/16 file called originalfile into a number of pieces.
> Recombined them and reversed the polarity with
> Sox inputfiles outputfile vol -1

Why did you cut it to pieces first?

> Then mixed the output file with the original with
> Sox -m originalfile outputfile combinedfile
> And then got stats for the combinedfile with
> Sox combinedfile -n stats

Why did you do all that?

> And I'm looking at pk level dB in the stats
> Although the combinedfile sounds perfectly nulled,
> stats reports pk level dB as -7

If you have a single nonzero sample at the level of -7dB,
then that's what your max peak is, even if the rest
of the file file "sounds nulled". Example:

$ sox -n file.wav synth 0.01 sin 440 gain -7
$ play file.wav stats

Sure "sounds nulled".


> Other files I've tried usually null to around -80
> I would expect nulls to -95 or so
> Any ideas why this file should sound ok but
> not get reported correctly in stats.
> Is it me reading the wrong parameter from stats,
> or using the wrong sox commands to accomplish this Nulling function

It's mostly you not telling us what you are trying to do,
not quoting your actual commands, and not showing us the file.

On May 21 19:33:43, [hidden email] wrote:

> This What I get from stats in the terminal
>
>              Overall     Left      Right
> DC offset  -0.000000 -0.000000 -0.000000
> Min level  -0.440674 -0.400543 -0.440674
> Max level   0.440918  0.416229  0.440918
> Pk lev dB      -7.11     -7.61     -7.11
> RMS lev dB    -30.81    -30.20    -31.52
> RMS Pk dB     -15.61    -15.61    -16.20
> RMS Tr dB     -96.87    -96.84    -96.87
> Crest factor       -     13.47     16.60
> Flat factor     0.00      0.00      0.00
> Pk count           2         2         2

This says that the peak was only attained at two "occasions":
To quote the manpage:

         Pk count is the number of occasions (not the number of samples)
         that the signal attained either Min level, or Max level.

That leads me to _speculate_ that there is a very short (inaudible) burst
somewhere in your file, which can very well be null everywhere else.

> Bit-depth      15/16     15/16     15/16
> Num samples    29.2M
> Length s     661.240
> Scale max   1.000000
> Window s       0.050


------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Sox-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/sox-users
Reply | Threaded
Open this post in threaded view
|

Re: Nulling files

Mew, Peter
Hi Jan
I am testing the accuracy of this way of Nulling files, eventually I will be adding processing to the split files, but not at this stage.
Is there a better way of achieving this?
I have reduced the long pathnames of the various files, to make it easier to read.
I can post the file somewhere if that would help.
Thanks for the pointer to the pk count number, this is most helpfull
Thanks
-pm


Sent from my iPad

> On 22 May 2014, at 08:05, "Jan Stary" <[hidden email]> wrote:
>
>> On May 21 19:16:35, [hidden email] wrote:
>> I have done the following using sox
>> Split a 44/16 file called originalfile into a number of pieces.
>> Recombined them and reversed the polarity with
>> Sox inputfiles outputfile vol -1
>
> Why did you cut it to pieces first?
>
>> Then mixed the output file with the original with
>> Sox -m originalfile outputfile combinedfile
>> And then got stats for the combinedfile with
>> Sox combinedfile -n stats
>
> Why did you do all that?
>
>> And I'm looking at pk level dB in the stats
>> Although the combinedfile sounds perfectly nulled,
>> stats reports pk level dB as -7
>
> If you have a single nonzero sample at the level of -7dB,
> then that's what your max peak is, even if the rest
> of the file file "sounds nulled". Example:
>
> $ sox -n file.wav synth 0.01 sin 440 gain -7
> $ play file.wav stats
>
> Sure "sounds nulled".
>
>
>> Other files I've tried usually null to around -80
>> I would expect nulls to -95 or so
>> Any ideas why this file should sound ok but
>> not get reported correctly in stats.
>> Is it me reading the wrong parameter from stats,
>> or using the wrong sox commands to accomplish this Nulling function
>
> It's mostly you not telling us what you are trying to do,
> not quoting your actual commands, and not showing us the file.
>
>> On May 21 19:33:43, [hidden email] wrote:
>> This What I get from stats in the terminal
>>
>>             Overall     Left      Right
>> DC offset  -0.000000 -0.000000 -0.000000
>> Min level  -0.440674 -0.400543 -0.440674
>> Max level   0.440918  0.416229  0.440918
>> Pk lev dB      -7.11     -7.61     -7.11
>> RMS lev dB    -30.81    -30.20    -31.52
>> RMS Pk dB     -15.61    -15.61    -16.20
>> RMS Tr dB     -96.87    -96.84    -96.87
>> Crest factor       -     13.47     16.60
>> Flat factor     0.00      0.00      0.00
>> Pk count           2         2         2
>
> This says that the peak was only attained at two "occasions":
> To quote the manpage:
>
>     Pk count is the number of occasions (not the number of samples)
>     that the signal attained either Min level, or Max level.
>
> That leads me to _speculate_ that there is a very short (inaudible) burst
> somewhere in your file, which can very well be null everywhere else.
>
>> Bit-depth      15/16     15/16     15/16
>> Num samples    29.2M
>> Length s     661.240
>> Scale max   1.000000
>> Window s       0.050
>
>
> ------------------------------------------------------------------------------
> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
> Instantly run your Selenium tests across 300+ browser/OS combos.
> Get unparalleled scalability from the best Selenium testing platform available
> Simple to use. Nothing to install. Get started now for free."
> http://p.sf.net/sfu/SauceLabs
> _______________________________________________
> Sox-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/sox-users

________________________________

“**Confidentiality**
The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.

We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Sox-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/sox-users
Reply | Threaded
Open this post in threaded view
|

R: Nulling files

Marco Bandiera
Please, could you explain me why bit depth from "stats" is 15/16 instead of 16/16?
I found the same in a channel of some stereo files recorded at 44,1kHz-16bit.
Thank you very much.

Marco

-----Messaggio originale-----
Da: Mew, Pete [mailto:[hidden email]]
Inviato: giovedì 22 maggio 2014 12:12
A: <[hidden email]>
Oggetto: Re: [SoX-users] Nulling files

Hi Jan
I am testing the accuracy of this way of Nulling files, eventually I will be adding processing to the split files, but not at this stage.
Is there a better way of achieving this?
I have reduced the long pathnames of the various files, to make it easier to read.
I can post the file somewhere if that would help.
Thanks for the pointer to the pk count number, this is most helpfull Thanks -pm


Sent from my iPad

> On 22 May 2014, at 08:05, "Jan Stary" <[hidden email]> wrote:
>
>> On May 21 19:16:35, [hidden email] wrote:
>> I have done the following using sox
>> Split a 44/16 file called originalfile into a number of pieces.
>> Recombined them and reversed the polarity with Sox inputfiles
>> outputfile vol -1
>
> Why did you cut it to pieces first?
>
>> Then mixed the output file with the original with Sox -m originalfile
>> outputfile combinedfile And then got stats for the combinedfile with
>> Sox combinedfile -n stats
>
> Why did you do all that?
>
>> And I'm looking at pk level dB in the stats Although the combinedfile
>> sounds perfectly nulled, stats reports pk level dB as -7
>
> If you have a single nonzero sample at the level of -7dB, then that's
> what your max peak is, even if the rest of the file file "sounds
> nulled". Example:
>
> $ sox -n file.wav synth 0.01 sin 440 gain -7 $ play file.wav stats
>
> Sure "sounds nulled".
>
>
>> Other files I've tried usually null to around -80 I would expect
>> nulls to -95 or so Any ideas why this file should sound ok but not
>> get reported correctly in stats.
>> Is it me reading the wrong parameter from stats, or using the wrong
>> sox commands to accomplish this Nulling function
>
> It's mostly you not telling us what you are trying to do, not quoting
> your actual commands, and not showing us the file.
>
>> On May 21 19:33:43, [hidden email] wrote:
>> This What I get from stats in the terminal
>>
>>             Overall     Left      Right
>> DC offset  -0.000000 -0.000000 -0.000000 Min level  -0.440674
>> -0.400543 -0.440674
>> Max level   0.440918  0.416229  0.440918
>> Pk lev dB      -7.11     -7.61     -7.11
>> RMS lev dB    -30.81    -30.20    -31.52
>> RMS Pk dB     -15.61    -15.61    -16.20
>> RMS Tr dB     -96.87    -96.84    -96.87
>> Crest factor       -     13.47     16.60
>> Flat factor     0.00      0.00      0.00
>> Pk count           2         2         2
>
> This says that the peak was only attained at two "occasions":
> To quote the manpage:
>
>     Pk count is the number of occasions (not the number of samples)
>     that the signal attained either Min level, or Max level.
>
> That leads me to _speculate_ that there is a very short (inaudible)
> burst somewhere in your file, which can very well be null everywhere else.
>
>> Bit-depth      15/16     15/16     15/16
>> Num samples    29.2M
>> Length s     661.240
>> Scale max   1.000000
>> Window s       0.050
>
>
> ----------------------------------------------------------------------
> -------- "Accelerate Dev Cycles with Automated Cross-Browser Testing -
> For FREE Instantly run your Selenium tests across 300+ browser/OS
> combos.
> Get unparalleled scalability from the best Selenium testing platform
> available Simple to use. Nothing to install. Get started now for free."
> http://p.sf.net/sfu/SauceLabs
> _______________________________________________
> Sox-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/sox-users

________________________________

“**Confidentiality**
The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.

We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Sox-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/sox-users


------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Sox-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/sox-users
Reply | Threaded
Open this post in threaded view
|

Re: R: Nulling files

Schumacher Marlon
Hi,

I suppose this is because of the peak level of -7dB (which leaves a little more than one bit of headroom in the file, thus 15 of 16 bits are used).

What puzzles me is that regardless of what Pete's contents of the input file was, shouldn't a reversal of polarity when mixed with the original _in any case_ result in only 'zero' samples? In other words, I don't understand how there can still be 2 peaks in the output file when mixing an input file with a polarity-reversed copy of itself?

Cheers,
Marlon

On May 22, 2014, at 09:59 , Marco Bandiera <[hidden email]> wrote:

> Please, could you explain me why bit depth from "stats" is 15/16 instead of 16/16?
> I found the same in a channel of some stereo files recorded at 44,1kHz-16bit.
> Thank you very much.
>
> Marco
>
> -----Messaggio originale-----
> Da: Mew, Pete [mailto:[hidden email]]
> Inviato: giovedì 22 maggio 2014 12:12
> A: <[hidden email]>
> Oggetto: Re: [SoX-users] Nulling files
>
> Hi Jan
> I am testing the accuracy of this way of Nulling files, eventually I will be adding processing to the split files, but not at this stage.
> Is there a better way of achieving this?
> I have reduced the long pathnames of the various files, to make it easier to read.
> I can post the file somewhere if that would help.
> Thanks for the pointer to the pk count number, this is most helpfull Thanks -pm
>
>
> Sent from my iPad
>
>> On 22 May 2014, at 08:05, "Jan Stary" <[hidden email]> wrote:
>>
>>> On May 21 19:16:35, [hidden email] wrote:
>>> I have done the following using sox
>>> Split a 44/16 file called originalfile into a number of pieces.
>>> Recombined them and reversed the polarity with Sox inputfiles
>>> outputfile vol -1
>>
>> Why did you cut it to pieces first?
>>
>>> Then mixed the output file with the original with Sox -m originalfile
>>> outputfile combinedfile And then got stats for the combinedfile with
>>> Sox combinedfile -n stats
>>
>> Why did you do all that?
>>
>>> And I'm looking at pk level dB in the stats Although the combinedfile
>>> sounds perfectly nulled, stats reports pk level dB as -7
>>
>> If you have a single nonzero sample at the level of -7dB, then that's
>> what your max peak is, even if the rest of the file file "sounds
>> nulled". Example:
>>
>> $ sox -n file.wav synth 0.01 sin 440 gain -7 $ play file.wav stats
>>
>> Sure "sounds nulled".
>>
>>
>>> Other files I've tried usually null to around -80 I would expect
>>> nulls to -95 or so Any ideas why this file should sound ok but not
>>> get reported correctly in stats.
>>> Is it me reading the wrong parameter from stats, or using the wrong
>>> sox commands to accomplish this Nulling function
>>
>> It's mostly you not telling us what you are trying to do, not quoting
>> your actual commands, and not showing us the file.
>>
>>> On May 21 19:33:43, [hidden email] wrote:
>>> This What I get from stats in the terminal
>>>
>>>           Overall     Left      Right
>>> DC offset  -0.000000 -0.000000 -0.000000 Min level  -0.440674
>>> -0.400543 -0.440674
>>> Max level   0.440918  0.416229  0.440918
>>> Pk lev dB      -7.11     -7.61     -7.11
>>> RMS lev dB    -30.81    -30.20    -31.52
>>> RMS Pk dB     -15.61    -15.61    -16.20
>>> RMS Tr dB     -96.87    -96.84    -96.87
>>> Crest factor       -     13.47     16.60
>>> Flat factor     0.00      0.00      0.00
>>> Pk count           2         2         2
>>
>> This says that the peak was only attained at two "occasions":
>> To quote the manpage:
>>
>>   Pk count is the number of occasions (not the number of samples)
>>   that the signal attained either Min level, or Max level.
>>
>> That leads me to _speculate_ that there is a very short (inaudible)
>> burst somewhere in your file, which can very well be null everywhere else.
>>
>>> Bit-depth      15/16     15/16     15/16
>>> Num samples    29.2M
>>> Length s     661.240
>>> Scale max   1.000000
>>> Window s       0.050
>>
>>
>> ----------------------------------------------------------------------
>> -------- "Accelerate Dev Cycles with Automated Cross-Browser Testing -
>> For FREE Instantly run your Selenium tests across 300+ browser/OS
>> combos.
>> Get unparalleled scalability from the best Selenium testing platform
>> available Simple to use. Nothing to install. Get started now for free."
>> http://p.sf.net/sfu/SauceLabs
>> _______________________________________________
>> Sox-users mailing list
>> [hidden email]
>> https://lists.sourceforge.net/lists/listinfo/sox-users
>
> ________________________________
>
> “**Confidentiality**
> The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.
>
> We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.
> ------------------------------------------------------------------------------
> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE Instantly run your Selenium tests across 300+ browser/OS combos.
> Get unparalleled scalability from the best Selenium testing platform available Simple to use. Nothing to install. Get started now for free."
> http://p.sf.net/sfu/SauceLabs
> _______________________________________________
> Sox-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/sox-users
>
>
> ------------------------------------------------------------------------------
> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
> Instantly run your Selenium tests across 300+ browser/OS combos.
> Get unparalleled scalability from the best Selenium testing platform available
> Simple to use. Nothing to install. Get started now for free."
> http://p.sf.net/sfu/SauceLabs
> _______________________________________________
> Sox-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/sox-users
>
>


------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Sox-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/sox-users
Reply | Threaded
Open this post in threaded view
|

Re: R: Nulling files

Mew, Peter
Hence my question.
More experimentation has shown that, as usual, sox is correct and I am not.
It appears that during the splitting and recombining one of the 16 bit files was forced to 24bit, and the other not. This, and other poor user implementation has caused some of the anomalies. I think this must have been a rogue file. I can now get nulls down -90 with most files, I'm looking at what's happening with the last bit

-pm

Sent from my iPad

> On 22 May 2014, at 17:19, "Schumacher Marlon" <[hidden email]> wrote:
>
> Hi,
>
> I suppose this is because of the peak level of -7dB (which leaves a little more than one bit of headroom in the file, thus 15 of 16 bits are used).
>
> What puzzles me is that regardless of what Pete's contents of the input file was, shouldn't a reversal of polarity when mixed with the original _in any case_ result in only 'zero' samples? In other words, I don't understand how there can still be 2 peaks in the output file when mixing an input file with a polarity-reversed copy of itself?
>
> Cheers,
> Marlon
>
>> On May 22, 2014, at 09:59 , Marco Bandiera <[hidden email]> wrote:
>>
>> Please, could you explain me why bit depth from "stats" is 15/16 instead of 16/16?
>> I found the same in a channel of some stereo files recorded at 44,1kHz-16bit.
>> Thank you very much.
>>
>> Marco
>>
>> -----Messaggio originale-----
>> Da: Mew, Pete [mailto:[hidden email]]
>> Inviato: giovedì 22 maggio 2014 12:12
>> A: <[hidden email]>
>> Oggetto: Re: [SoX-users] Nulling files
>>
>> Hi Jan
>> I am testing the accuracy of this way of Nulling files, eventually I will be adding processing to the split files, but not at this stage.
>> Is there a better way of achieving this?
>> I have reduced the long pathnames of the various files, to make it easier to read.
>> I can post the file somewhere if that would help.
>> Thanks for the pointer to the pk count number, this is most helpfull Thanks -pm
>>
>>
>> Sent from my iPad
>>
>>>> On 22 May 2014, at 08:05, "Jan Stary" <[hidden email]> wrote:
>>>>
>>>> On May 21 19:16:35, [hidden email] wrote:
>>>> I have done the following using sox
>>>> Split a 44/16 file called originalfile into a number of pieces.
>>>> Recombined them and reversed the polarity with Sox inputfiles
>>>> outputfile vol -1
>>>
>>> Why did you cut it to pieces first?
>>>
>>>> Then mixed the output file with the original with Sox -m originalfile
>>>> outputfile combinedfile And then got stats for the combinedfile with
>>>> Sox combinedfile -n stats
>>>
>>> Why did you do all that?
>>>
>>>> And I'm looking at pk level dB in the stats Although the combinedfile
>>>> sounds perfectly nulled, stats reports pk level dB as -7
>>>
>>> If you have a single nonzero sample at the level of -7dB, then that's
>>> what your max peak is, even if the rest of the file file "sounds
>>> nulled". Example:
>>>
>>> $ sox -n file.wav synth 0.01 sin 440 gain -7 $ play file.wav stats
>>>
>>> Sure "sounds nulled".
>>>
>>>
>>>> Other files I've tried usually null to around -80 I would expect
>>>> nulls to -95 or so Any ideas why this file should sound ok but not
>>>> get reported correctly in stats.
>>>> Is it me reading the wrong parameter from stats, or using the wrong
>>>> sox commands to accomplish this Nulling function
>>>
>>> It's mostly you not telling us what you are trying to do, not quoting
>>> your actual commands, and not showing us the file.
>>>
>>>> On May 21 19:33:43, [hidden email] wrote:
>>>> This What I get from stats in the terminal
>>>>
>>>>          Overall     Left      Right
>>>> DC offset  -0.000000 -0.000000 -0.000000 Min level  -0.440674
>>>> -0.400543 -0.440674
>>>> Max level   0.440918  0.416229  0.440918
>>>> Pk lev dB      -7.11     -7.61     -7.11
>>>> RMS lev dB    -30.81    -30.20    -31.52
>>>> RMS Pk dB     -15.61    -15.61    -16.20
>>>> RMS Tr dB     -96.87    -96.84    -96.87
>>>> Crest factor       -     13.47     16.60
>>>> Flat factor     0.00      0.00      0.00
>>>> Pk count           2         2         2
>>>
>>> This says that the peak was only attained at two "occasions":
>>> To quote the manpage:
>>>
>>>  Pk count is the number of occasions (not the number of samples)
>>>  that the signal attained either Min level, or Max level.
>>>
>>> That leads me to _speculate_ that there is a very short (inaudible)
>>> burst somewhere in your file, which can very well be null everywhere else.
>>>
>>>> Bit-depth      15/16     15/16     15/16
>>>> Num samples    29.2M
>>>> Length s     661.240
>>>> Scale max   1.000000
>>>> Window s       0.050
>>>
>>>
>>> ----------------------------------------------------------------------
>>> -------- "Accelerate Dev Cycles with Automated Cross-Browser Testing -
>>> For FREE Instantly run your Selenium tests across 300+ browser/OS
>>> combos.
>>> Get unparalleled scalability from the best Selenium testing platform
>>> available Simple to use. Nothing to install. Get started now for free."
>>> http://p.sf.net/sfu/SauceLabs
>>> _______________________________________________
>>> Sox-users mailing list
>>> [hidden email]
>>> https://lists.sourceforge.net/lists/listinfo/sox-users
>>
>> ________________________________
>>
>> “**Confidentiality**
>> The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.
>>
>> We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.
>> ------------------------------------------------------------------------------
>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE Instantly run your Selenium tests across 300+ browser/OS combos.
>> Get unparalleled scalability from the best Selenium testing platform available Simple to use. Nothing to install. Get started now for free."
>> http://p.sf.net/sfu/SauceLabs
>> _______________________________________________
>> Sox-users mailing list
>> [hidden email]
>> https://lists.sourceforge.net/lists/listinfo/sox-users
>>
>>
>> ------------------------------------------------------------------------------
>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
>> Instantly run your Selenium tests across 300+ browser/OS combos.
>> Get unparalleled scalability from the best Selenium testing platform available
>> Simple to use. Nothing to install. Get started now for free."
>> http://p.sf.net/sfu/SauceLabs
>> _______________________________________________
>> Sox-users mailing list
>> [hidden email]
>> https://lists.sourceforge.net/lists/listinfo/sox-users
>
>
> ------------------------------------------------------------------------------
> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
> Instantly run your Selenium tests across 300+ browser/OS combos.
> Get unparalleled scalability from the best Selenium testing platform available
> Simple to use. Nothing to install. Get started now for free."
> http://p.sf.net/sfu/SauceLabs
> _______________________________________________
> Sox-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/sox-users

________________________________

“**Confidentiality**
The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.

We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.

------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Sox-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/sox-users
Reply | Threaded
Open this post in threaded view
|

Re: Nulling files

Jan Stary
In reply to this post by Schumacher Marlon
On May 22 11:45:07, [hidden email] wrote:
> What puzzles me is that regardless of what Pete's contents of the input file was, shouldn't a reversal of polarity when mixed with the original _in any case_ result in only 'zero' samples? In other words, I don't understand how there can still be 2 peaks in the output file when mixing an input file with a polarity-reversed copy of itself?

It's _not_ a polarity reversed copy. He "split it to pieces
and then recombined them", in an unspecified way, and only
then reversed the polarity.


------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Sox-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/sox-users
Reply | Threaded
Open this post in threaded view
|

Re: Nulling files

Mew, Peter
Not unspecified.
I joined them using sox
-pm

Sent from my iPad

> On 23 May 2014, at 12:05, "Jan Stary" <[hidden email]> wrote:
>
>> On May 22 11:45:07, [hidden email] wrote:
>> What puzzles me is that regardless of what Pete's contents of the input file was, shouldn't a reversal of polarity when mixed with the original _in any case_ result in only 'zero' samples? In other words, I don't understand how there can still be 2 peaks in the output file when mixing an input file with a polarity-reversed copy of itself?
>
> It's _not_ a polarity reversed copy. He "split it to pieces
> and then recombined them", in an unspecified way, and only
> then reversed the polarity.
>
>
> ------------------------------------------------------------------------------
> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
> Instantly run your Selenium tests across 300+ browser/OS combos.
> Get unparalleled scalability from the best Selenium testing platform available
> Simple to use. Nothing to install. Get started now for free."
> http://p.sf.net/sfu/SauceLabs
> _______________________________________________
> Sox-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/sox-users

________________________________

“**Confidentiality**
The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.

We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Sox-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/sox-users
Reply | Threaded
Open this post in threaded view
|

Re: Nulling files

Schumacher Marlon
In reply to this post by Mew, Peter
Hi,

Reading over this again I got curious:
"This, and other poor user implementation has caused some of the anomalies. I think this must have been a rogue file. "
-> could you precise this a bit (poor user implementation)?

Also, I wonder how come the nulled files have a peak level of -90dB (i.e. why this number - which roughly corresponds to 0.000032 in normalized amplitude)? Shouldn't it be -inf dB? Or depending on the bit depth of the file the smallest possible value, e.g. 24Bit -> 144dB?

Best,
Marlon

On May 22, 2014, at 17:45 , Mew, Pete <[hidden email]> wrote:

> Hence my question.
> More experimentation has shown that, as usual, sox is correct and I am not.
> It appears that during the splitting and recombining one of the 16 bit files was forced to 24bit, and the other not. This, and other poor user implementation has caused some of the anomalies. I think this must have been a rogue file. I can now get nulls down -90 with most files, I'm looking at what's happening with the last bit
>
> -pm
>
> Sent from my iPad
>
>> On 22 May 2014, at 17:19, "Schumacher Marlon" <[hidden email]> wrote:
>>
>> Hi,
>>
>> I suppose this is because of the peak level of -7dB (which leaves a little more than one bit of headroom in the file, thus 15 of 16 bits are used).
>>
>> What puzzles me is that regardless of what Pete's contents of the input file was, shouldn't a reversal of polarity when mixed with the original _in any case_ result in only 'zero' samples? In other words, I don't understand how there can still be 2 peaks in the output file when mixing an input file with a polarity-reversed copy of itself?
>>
>> Cheers,
>> Marlon
>>
>>> On May 22, 2014, at 09:59 , Marco Bandiera <[hidden email]> wrote:
>>>
>>> Please, could you explain me why bit depth from "stats" is 15/16 instead of 16/16?
>>> I found the same in a channel of some stereo files recorded at 44,1kHz-16bit.
>>> Thank you very much.
>>>
>>> Marco
>>>
>>> -----Messaggio originale-----
>>> Da: Mew, Pete [mailto:[hidden email]]
>>> Inviato: giovedì 22 maggio 2014 12:12
>>> A: <[hidden email]>
>>> Oggetto: Re: [SoX-users] Nulling files
>>>
>>> Hi Jan
>>> I am testing the accuracy of this way of Nulling files, eventually I will be adding processing to the split files, but not at this stage.
>>> Is there a better way of achieving this?
>>> I have reduced the long pathnames of the various files, to make it easier to read.
>>> I can post the file somewhere if that would help.
>>> Thanks for the pointer to the pk count number, this is most helpfull Thanks -pm
>>>
>>>
>>> Sent from my iPad
>>>
>>>>> On 22 May 2014, at 08:05, "Jan Stary" <[hidden email]> wrote:
>>>>>
>>>>> On May 21 19:16:35, [hidden email] wrote:
>>>>> I have done the following using sox
>>>>> Split a 44/16 file called originalfile into a number of pieces.
>>>>> Recombined them and reversed the polarity with Sox inputfiles
>>>>> outputfile vol -1
>>>>
>>>> Why did you cut it to pieces first?
>>>>
>>>>> Then mixed the output file with the original with Sox -m originalfile
>>>>> outputfile combinedfile And then got stats for the combinedfile with
>>>>> Sox combinedfile -n stats
>>>>
>>>> Why did you do all that?
>>>>
>>>>> And I'm looking at pk level dB in the stats Although the combinedfile
>>>>> sounds perfectly nulled, stats reports pk level dB as -7
>>>>
>>>> If you have a single nonzero sample at the level of -7dB, then that's
>>>> what your max peak is, even if the rest of the file file "sounds
>>>> nulled". Example:
>>>>
>>>> $ sox -n file.wav synth 0.01 sin 440 gain -7 $ play file.wav stats
>>>>
>>>> Sure "sounds nulled".
>>>>
>>>>
>>>>> Other files I've tried usually null to around -80 I would expect
>>>>> nulls to -95 or so Any ideas why this file should sound ok but not
>>>>> get reported correctly in stats.
>>>>> Is it me reading the wrong parameter from stats, or using the wrong
>>>>> sox commands to accomplish this Nulling function
>>>>
>>>> It's mostly you not telling us what you are trying to do, not quoting
>>>> your actual commands, and not showing us the file.
>>>>
>>>>> On May 21 19:33:43, [hidden email] wrote:
>>>>> This What I get from stats in the terminal
>>>>>
>>>>>         Overall     Left      Right
>>>>> DC offset  -0.000000 -0.000000 -0.000000 Min level  -0.440674
>>>>> -0.400543 -0.440674
>>>>> Max level   0.440918  0.416229  0.440918
>>>>> Pk lev dB      -7.11     -7.61     -7.11
>>>>> RMS lev dB    -30.81    -30.20    -31.52
>>>>> RMS Pk dB     -15.61    -15.61    -16.20
>>>>> RMS Tr dB     -96.87    -96.84    -96.87
>>>>> Crest factor       -     13.47     16.60
>>>>> Flat factor     0.00      0.00      0.00
>>>>> Pk count           2         2         2
>>>>
>>>> This says that the peak was only attained at two "occasions":
>>>> To quote the manpage:
>>>>
>>>> Pk count is the number of occasions (not the number of samples)
>>>> that the signal attained either Min level, or Max level.
>>>>
>>>> That leads me to _speculate_ that there is a very short (inaudible)
>>>> burst somewhere in your file, which can very well be null everywhere else.
>>>>
>>>>> Bit-depth      15/16     15/16     15/16
>>>>> Num samples    29.2M
>>>>> Length s     661.240
>>>>> Scale max   1.000000
>>>>> Window s       0.050
>>>>
>>>>
>>>> ----------------------------------------------------------------------
>>>> -------- "Accelerate Dev Cycles with Automated Cross-Browser Testing -
>>>> For FREE Instantly run your Selenium tests across 300+ browser/OS
>>>> combos.
>>>> Get unparalleled scalability from the best Selenium testing platform
>>>> available Simple to use. Nothing to install. Get started now for free."
>>>> http://p.sf.net/sfu/SauceLabs
>>>> _______________________________________________
>>>> Sox-users mailing list
>>>> [hidden email]
>>>> https://lists.sourceforge.net/lists/listinfo/sox-users
>>>
>>> ________________________________
>>>
>>> “**Confidentiality**
>>> The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.
>>>
>>> We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.
>>> ------------------------------------------------------------------------------
>>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE Instantly run your Selenium tests across 300+ browser/OS combos.
>>> Get unparalleled scalability from the best Selenium testing platform available Simple to use. Nothing to install. Get started now for free."
>>> http://p.sf.net/sfu/SauceLabs
>>> _______________________________________________
>>> Sox-users mailing list
>>> [hidden email]
>>> https://lists.sourceforge.net/lists/listinfo/sox-users
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
>>> Instantly run your Selenium tests across 300+ browser/OS combos.
>>> Get unparalleled scalability from the best Selenium testing platform available
>>> Simple to use. Nothing to install. Get started now for free."
>>> http://p.sf.net/sfu/SauceLabs
>>> _______________________________________________
>>> Sox-users mailing list
>>> [hidden email]
>>> https://lists.sourceforge.net/lists/listinfo/sox-users
>>
>>
>> ------------------------------------------------------------------------------
>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
>> Instantly run your Selenium tests across 300+ browser/OS combos.
>> Get unparalleled scalability from the best Selenium testing platform available
>> Simple to use. Nothing to install. Get started now for free."
>> http://p.sf.net/sfu/SauceLabs
>> _______________________________________________
>> Sox-users mailing list
>> [hidden email]
>> https://lists.sourceforge.net/lists/listinfo/sox-users
>
> ________________________________
>
> “**Confidentiality**
> The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.
>
> We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.
>
> ------------------------------------------------------------------------------
> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
> Instantly run your Selenium tests across 300+ browser/OS combos.
> Get unparalleled scalability from the best Selenium testing platform available
> Simple to use. Nothing to install. Get started now for free."
> http://p.sf.net/sfu/SauceLabs
> _______________________________________________
> Sox-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/sox-users
>


------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Sox-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/sox-users
Reply | Threaded
Open this post in threaded view
|

Re: Nulling files

Jan Stary
In reply to this post by Mew, Peter
On May 25 22:45:12, [hidden email] wrote:
> Not unspecified.
> I joined them using sox

Oh, "using sox". Great, that explains it then.

What I mean by "unspecified" is that you don't give
your full and exact command lines, you only say
"I cut them to pieces" and so.


>
> Sent from my iPad
>
> > On 23 May 2014, at 12:05, "Jan Stary" <[hidden email]> wrote:
> >
> >> On May 22 11:45:07, [hidden email] wrote:
> >> What puzzles me is that regardless of what Pete's contents of the input file was, shouldn't a reversal of polarity when mixed with the original _in any case_ result in only 'zero' samples? In other words, I don't understand how there can still be 2 peaks in the output file when mixing an input file with a polarity-reversed copy of itself?
> >
> > It's _not_ a polarity reversed copy. He "split it to pieces
> > and then recombined them", in an unspecified way, and only
> > then reversed the polarity.
> >
> >
> > ------------------------------------------------------------------------------
> > "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
> > Instantly run your Selenium tests across 300+ browser/OS combos.
> > Get unparalleled scalability from the best Selenium testing platform available
> > Simple to use. Nothing to install. Get started now for free."
> > http://p.sf.net/sfu/SauceLabs
> > _______________________________________________
> > Sox-users mailing list
> > [hidden email]
> > https://lists.sourceforge.net/lists/listinfo/sox-users
>
> ________________________________
>
> ???**Confidentiality**
> The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.
>
> We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.
> ------------------------------------------------------------------------------
> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
> Instantly run your Selenium tests across 300+ browser/OS combos.
> Get unparalleled scalability from the best Selenium testing platform available
> Simple to use. Nothing to install. Get started now for free."
> http://p.sf.net/sfu/SauceLabs
> _______________________________________________
> Sox-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/sox-users

------------------------------------------------------------------------------
The best possible search technologies are now affordable for all companies.
Download your FREE open source Enterprise Search Engine today!
Our experts will assist you in its installation for $59/mo, no commitment.
Test it for FREE on our Cloud platform anytime!
http://pubads.g.doubleclick.net/gampad/clk?id=145328191&iu=/4140/ostg.clktrk
_______________________________________________
Sox-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/sox-users
Reply | Threaded
Open this post in threaded view
|

Re: Nulling files

Mew, Peter
In reply to this post by Schumacher Marlon
Poor user implementation = I was not as rigorous as I should have been setting up the tests.
-90 is roughly equal to 15 bits, I therefore suspect that there is still some dither going on somewhere
Previous tests I have done show that 24 bit files will usually null to - inf but 16 bit files will only null to about -90. Again, I suspect I have not switched off all the dither settings during the tests.
If I understand correctly, sox converts everything to 32 bit before invoking any effects, if -D is not selected, does sox apply dither when converting back to 16 bit
Cheers
-pm

Sent from my iPad

> On 26 May 2014, at 00:06, "Schumacher Marlon" <[hidden email]> wrote:
>
> Hi,
>
> Reading over this again I got curious:
> "This, and other poor user implementation has caused some of the anomalies. I think this must have been a rogue file. "
> -> could you precise this a bit (poor user implementation)?
>
> Also, I wonder how come the nulled files have a peak level of -90dB (i.e. why this number - which roughly corresponds to 0.000032 in normalized amplitude)? Shouldn't it be -inf dB? Or depending on the bit depth of the file the smallest possible value, e.g. 24Bit -> 144dB?
>
> Best,
> Marlon
>
>> On May 22, 2014, at 17:45 , Mew, Pete <[hidden email]> wrote:
>>
>> Hence my question.
>> More experimentation has shown that, as usual, sox is correct and I am not.
>> It appears that during the splitting and recombining one of the 16 bit files was forced to 24bit, and the other not. This, and other poor user implementation has caused some of the anomalies. I think this must have been a rogue file. I can now get nulls down -90 with most files, I'm looking at what's happening with the last bit
>>
>> -pm
>>
>> Sent from my iPad
>>
>>> On 22 May 2014, at 17:19, "Schumacher Marlon" <[hidden email]> wrote:
>>>
>>> Hi,
>>>
>>> I suppose this is because of the peak level of -7dB (which leaves a little more than one bit of headroom in the file, thus 15 of 16 bits are used).
>>>
>>> What puzzles me is that regardless of what Pete's contents of the input file was, shouldn't a reversal of polarity when mixed with the original _in any case_ result in only 'zero' samples? In other words, I don't understand how there can still be 2 peaks in the output file when mixing an input file with a polarity-reversed copy of itself?
>>>
>>> Cheers,
>>> Marlon
>>>
>>>> On May 22, 2014, at 09:59 , Marco Bandiera <[hidden email]> wrote:
>>>>
>>>> Please, could you explain me why bit depth from "stats" is 15/16 instead of 16/16?
>>>> I found the same in a channel of some stereo files recorded at 44,1kHz-16bit.
>>>> Thank you very much.
>>>>
>>>> Marco
>>>>
>>>> -----Messaggio originale-----
>>>> Da: Mew, Pete [mailto:[hidden email]]
>>>> Inviato: giovedì 22 maggio 2014 12:12
>>>> A: <[hidden email]>
>>>> Oggetto: Re: [SoX-users] Nulling files
>>>>
>>>> Hi Jan
>>>> I am testing the accuracy of this way of Nulling files, eventually I will be adding processing to the split files, but not at this stage.
>>>> Is there a better way of achieving this?
>>>> I have reduced the long pathnames of the various files, to make it easier to read.
>>>> I can post the file somewhere if that would help.
>>>> Thanks for the pointer to the pk count number, this is most helpfull Thanks -pm
>>>>
>>>>
>>>> Sent from my iPad
>>>>
>>>>>> On 22 May 2014, at 08:05, "Jan Stary" <[hidden email]> wrote:
>>>>>>
>>>>>> On May 21 19:16:35, [hidden email] wrote:
>>>>>> I have done the following using sox
>>>>>> Split a 44/16 file called originalfile into a number of pieces.
>>>>>> Recombined them and reversed the polarity with Sox inputfiles
>>>>>> outputfile vol -1
>>>>>
>>>>> Why did you cut it to pieces first?
>>>>>
>>>>>> Then mixed the output file with the original with Sox -m originalfile
>>>>>> outputfile combinedfile And then got stats for the combinedfile with
>>>>>> Sox combinedfile -n stats
>>>>>
>>>>> Why did you do all that?
>>>>>
>>>>>> And I'm looking at pk level dB in the stats Although the combinedfile
>>>>>> sounds perfectly nulled, stats reports pk level dB as -7
>>>>>
>>>>> If you have a single nonzero sample at the level of -7dB, then that's
>>>>> what your max peak is, even if the rest of the file file "sounds
>>>>> nulled". Example:
>>>>>
>>>>> $ sox -n file.wav synth 0.01 sin 440 gain -7 $ play file.wav stats
>>>>>
>>>>> Sure "sounds nulled".
>>>>>
>>>>>
>>>>>> Other files I've tried usually null to around -80 I would expect
>>>>>> nulls to -95 or so Any ideas why this file should sound ok but not
>>>>>> get reported correctly in stats.
>>>>>> Is it me reading the wrong parameter from stats, or using the wrong
>>>>>> sox commands to accomplish this Nulling function
>>>>>
>>>>> It's mostly you not telling us what you are trying to do, not quoting
>>>>> your actual commands, and not showing us the file.
>>>>>
>>>>>> On May 21 19:33:43, [hidden email] wrote:
>>>>>> This What I get from stats in the terminal
>>>>>>
>>>>>>        Overall     Left      Right
>>>>>> DC offset  -0.000000 -0.000000 -0.000000 Min level  -0.440674
>>>>>> -0.400543 -0.440674
>>>>>> Max level   0.440918  0.416229  0.440918
>>>>>> Pk lev dB      -7.11     -7.61     -7.11
>>>>>> RMS lev dB    -30.81    -30.20    -31.52
>>>>>> RMS Pk dB     -15.61    -15.61    -16.20
>>>>>> RMS Tr dB     -96.87    -96.84    -96.87
>>>>>> Crest factor       -     13.47     16.60
>>>>>> Flat factor     0.00      0.00      0.00
>>>>>> Pk count           2         2         2
>>>>>
>>>>> This says that the peak was only attained at two "occasions":
>>>>> To quote the manpage:
>>>>>
>>>>> Pk count is the number of occasions (not the number of samples)
>>>>> that the signal attained either Min level, or Max level.
>>>>>
>>>>> That leads me to _speculate_ that there is a very short (inaudible)
>>>>> burst somewhere in your file, which can very well be null everywhere else.
>>>>>
>>>>>> Bit-depth      15/16     15/16     15/16
>>>>>> Num samples    29.2M
>>>>>> Length s     661.240
>>>>>> Scale max   1.000000
>>>>>> Window s       0.050
>>>>>
>>>>>
>>>>> ----------------------------------------------------------------------
>>>>> -------- "Accelerate Dev Cycles with Automated Cross-Browser Testing -
>>>>> For FREE Instantly run your Selenium tests across 300+ browser/OS
>>>>> combos.
>>>>> Get unparalleled scalability from the best Selenium testing platform
>>>>> available Simple to use. Nothing to install. Get started now for free."
>>>>> http://p.sf.net/sfu/SauceLabs
>>>>> _______________________________________________
>>>>> Sox-users mailing list
>>>>> [hidden email]
>>>>> https://lists.sourceforge.net/lists/listinfo/sox-users
>>>>
>>>> ________________________________
>>>>
>>>> “**Confidentiality**
>>>> The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.
>>>>
>>>> We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.
>>>> ------------------------------------------------------------------------------
>>>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE Instantly run your Selenium tests across 300+ browser/OS combos.
>>>> Get unparalleled scalability from the best Selenium testing platform available Simple to use. Nothing to install. Get started now for free."
>>>> http://p.sf.net/sfu/SauceLabs
>>>> _______________________________________________
>>>> Sox-users mailing list
>>>> [hidden email]
>>>> https://lists.sourceforge.net/lists/listinfo/sox-users
>>>>
>>>>
>>>> ------------------------------------------------------------------------------
>>>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
>>>> Instantly run your Selenium tests across 300+ browser/OS combos.
>>>> Get unparalleled scalability from the best Selenium testing platform available
>>>> Simple to use. Nothing to install. Get started now for free."
>>>> http://p.sf.net/sfu/SauceLabs
>>>> _______________________________________________
>>>> Sox-users mailing list
>>>> [hidden email]
>>>> https://lists.sourceforge.net/lists/listinfo/sox-users
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
>>> Instantly run your Selenium tests across 300+ browser/OS combos.
>>> Get unparalleled scalability from the best Selenium testing platform available
>>> Simple to use. Nothing to install. Get started now for free."
>>> http://p.sf.net/sfu/SauceLabs
>>> _______________________________________________
>>> Sox-users mailing list
>>> [hidden email]
>>> https://lists.sourceforge.net/lists/listinfo/sox-users
>>
>> ________________________________
>>
>> “**Confidentiality**
>> The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.
>>
>> We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.
>>
>> ------------------------------------------------------------------------------
>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
>> Instantly run your Selenium tests across 300+ browser/OS combos.
>> Get unparalleled scalability from the best Selenium testing platform available
>> Simple to use. Nothing to install. Get started now for free."
>> http://p.sf.net/sfu/SauceLabs
>> _______________________________________________
>> Sox-users mailing list
>> [hidden email]
>> https://lists.sourceforge.net/lists/listinfo/sox-users
>
>
> ------------------------------------------------------------------------------
> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
> Instantly run your Selenium tests across 300+ browser/OS combos.
> Get unparalleled scalability from the best Selenium testing platform available
> Simple to use. Nothing to install. Get started now for free."
> http://p.sf.net/sfu/SauceLabs
> _______________________________________________
> Sox-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/sox-users

________________________________

“**Confidentiality**
The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.

We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.
------------------------------------------------------------------------------
The best possible search technologies are now affordable for all companies.
Download your FREE open source Enterprise Search Engine today!
Our experts will assist you in its installation for $59/mo, no commitment.
Test it for FREE on our Cloud platform anytime!
http://pubads.g.doubleclick.net/gampad/clk?id=145328191&iu=/4140/ostg.clktrk
_______________________________________________
Sox-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/sox-users
Reply | Threaded
Open this post in threaded view
|

Re: Nulling files

Mew, Peter
In reply to this post by Jan Stary
Sox piece1.wav piece2.wav piece3.wav....... Piecen.wav out file.wav
-pm

Sent from my iPad

> On 26 May 2014, at 09:31, "Jan Stary" <[hidden email]> wrote:
>
>> On May 25 22:45:12, [hidden email] wrote:
>> Not unspecified.
>> I joined them using sox
>
> Oh, "using sox". Great, that explains it then.
>
> What I mean by "unspecified" is that you don't give
> your full and exact command lines, you only say
> "I cut them to pieces" and so.
>
>
>>
>> Sent from my iPad
>>
>>>> On 23 May 2014, at 12:05, "Jan Stary" <[hidden email]> wrote:
>>>>
>>>> On May 22 11:45:07, [hidden email] wrote:
>>>> What puzzles me is that regardless of what Pete's contents of the input file was, shouldn't a reversal of polarity when mixed with the original _in any case_ result in only 'zero' samples? In other words, I don't understand how there can still be 2 peaks in the output file when mixing an input file with a polarity-reversed copy of itself?
>>>
>>> It's _not_ a polarity reversed copy. He "split it to pieces
>>> and then recombined them", in an unspecified way, and only
>>> then reversed the polarity.
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
>>> Instantly run your Selenium tests across 300+ browser/OS combos.
>>> Get unparalleled scalability from the best Selenium testing platform available
>>> Simple to use. Nothing to install. Get started now for free."
>>> http://p.sf.net/sfu/SauceLabs
>>> _______________________________________________
>>> Sox-users mailing list
>>> [hidden email]
>>> https://lists.sourceforge.net/lists/listinfo/sox-users
>>
>> ________________________________
>>
>> ???**Confidentiality**
>> The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.
>>
>> We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.
>> ------------------------------------------------------------------------------
>> "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
>> Instantly run your Selenium tests across 300+ browser/OS combos.
>> Get unparalleled scalability from the best Selenium testing platform available
>> Simple to use. Nothing to install. Get started now for free."
>> http://p.sf.net/sfu/SauceLabs
>> _______________________________________________
>> Sox-users mailing list
>> [hidden email]
>> https://lists.sourceforge.net/lists/listinfo/sox-users
>
> ------------------------------------------------------------------------------
> The best possible search technologies are now affordable for all companies.
> Download your FREE open source Enterprise Search Engine today!
> Our experts will assist you in its installation for $59/mo, no commitment.
> Test it for FREE on our Cloud platform anytime!
> http://pubads.g.doubleclick.net/gampad/clk?id=145328191&iu=/4140/ostg.clktrk
> _______________________________________________
> Sox-users mailing list
> [hidden email]
> https://lists.sourceforge.net/lists/listinfo/sox-users

________________________________

“**Confidentiality**
The information contained in this e-mail is confidential, may be privileged and is intended solely for the use of the named addressee. Access to this e-mail by any other person is not authorised. If you are not the intended recipient, you should not disclose, copy, distribute, take any action or rely on it and you should please notify the sender by reply. Any opinions expressed are not necessarily those of the company.

We may monitor all incoming and outgoing emails in line with current legislation. We have taken steps to ensure that this email and attachments are free from any virus, but it remains your responsibility to ensure that viruses do not adversely affect you.
------------------------------------------------------------------------------
The best possible search technologies are now affordable for all companies.
Download your FREE open source Enterprise Search Engine today!
Our experts will assist you in its installation for $59/mo, no commitment.
Test it for FREE on our Cloud platform anytime!
http://pubads.g.doubleclick.net/gampad/clk?id=145328191&iu=/4140/ostg.clktrk
_______________________________________________
Sox-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/sox-users