On 2019-10-30 14:42, Nils Wallgren wrote:
I downloaded it from sourceforge. Originally choose the sox
14.4.2-win32.exe which is an installer.
I tried the the sox-14-4-2-win32.zip but this I hade to drag to my
program files (x86) myself. I also had to a path so that my command
prompt recognises the sox.
So... you added to PATH the folder in which you placed sox.exe
and (later) play.exe, rec.exe and soxi.exe ?
When I evalutate sox I get:
SoX v14.4.2
Sox FAIL sox: Not enough input filenames specified
Usage summary: [gopts] ...
And bunch of other things that looks lika a manual
That's the expected reply if you have no parameters (or maybe
the wrong ones).
I tried sox -t waveaudio -d "C:\Users\xx\Documents\bla.wav"
And at least it created a file.
And did the file contain any audio - if eg you double-click
it did it play? I'd expect a wav file just to play on Windows,
- you don't need sox for that. Here, the default player for
wav files is Windows Media Player.
I also tried(after I created a copy of the sox.exe and named it
rec.exe, another copy play.exe and another soxi.exe):
rec new-file.wav ( an example taken from page 3 from the manual)
But this doesn’t create any file.
The manual says that: rec new-file.wav
is equivalent to: sox −d new-file.wav
ie to copy audio from -d which is the default device, to the named
file.
Are you sure it didn't create a file, but maybe put it somewhere that
you haven't looked? Why didn't you use eg
rec "C:\Users\xx\Documents\another.wav"
Also, you'll note that that command doesn't have: -t waveaudio
which the working one did have. Why did you miss it out?
I've never used sox to record audio from the same PC (I use it
to manipulate files made on a solid-state audio recorder), but
reading the soxformat manual suggests that this is maybe used
to tell sox which device to use. It looks as if you might need
to use
-t waveaudio "the device name"
or
-t waveaudio n
where n is a device number. (But later, trying to specify an
output device this way, I couldn't get anything to work, with
either n values 0,1,2,3,4... or names like "Speakers" or "USB
Audio CODEC" which are what Windows sound configuration shows
me. (I've got W8.1.)
Then again, you say that the earlier command did create a
bla.wav file. But was there anything in it?
You can get useful info from an audio file's headers (if it has
any headers) by eg:
soxi "C:\this\that\sounds.wav"
and also two of the "effects" that sox can run will provide info about
the audio itself inside the file. For exaample:
C:\>soxi "C:\Dropbox\JN_Recordings\20150615 SCC - Flickering Light\Raw
files\3 SV3900 - Concert.wav"
Input File : 'C:\Dropbox\JN_Recordings\20150615 SCC - Flickering
Light\Raw files\3 SV3900 - Concert.wav'
Channels : 2
Sample Rate : 44100
Precision : 16-bit
Duration : 00:42:10.84 = 111609856 samples = 189813 CDDA sectors
File Size : 446M
Bit Rate : 1.41M
Sample Encoding: 16-bit Signed Integer PCM
C:\>
The useful info effects are named "stat" and "stats". For them I get:
C:\>sox "C:\Dropbox\JN_Recordings\20150615 SCC - Flickering Light\Raw
files\3 SV3900 - Concert.wav" -n stat
Samples read: 223219712
Length (seconds): 2530.835737
Scaled by: 2147483647.0
Maximum amplitude: 0.418976
Minimum amplitude: -0.426361
Midline amplitude: -0.003693
Mean norm: 0.008174
Mean amplitude: 0.000096
RMS amplitude: 0.015213
Maximum delta: 0.362122
Minimum delta: 0.000000
Mean delta: 0.007552
RMS delta: 0.013745
Rough frequency: 6341
Volume adjustment: 2.345
C:\>sox "C:\Dropbox\JN_Recordings\20150615 SCC - Flickering Light\Raw
files\3 SV3900 - Concert.wav" -n stats
Overall Left Right
DC offset 0.000162 0.000162 0.000030
Min level -0.426361 -0.426361 -0.355652
Max level 0.418976 0.418976 0.331818
Pk lev dB -7.40 -7.40 -8.98
RMS lev dB -36.36 -35.29 -37.78
RMS Pk dB -16.18 -16.18 -16.96
RMS Tr dB -85.48 -77.99 -85.48
Crest factor - 24.78 27.54
Flat factor 0.00 0.00 0.00
Pk count 2 2 2
Bit-depth 15/16 15/16 15/16
Num samples 112M
Length s 2530.836
Scale max 1.000000
Window s 0.050
C:\>
Note that here sox reads audio from the named file, and copies it to
"-n"
which is a null device (ie it is copied to nowhere). However as the
data
is seen it is processed by the effect.
You can do that command as: sox "inputfile.wav" -n stat stats
torun both "effects" one after the other if you prefer.
Also tried this : play -n -c1 synth sin %-12 sin %-9 sin %-5 sin %-2
fade h 0.1 1 0.1 from from the manual but no luck
No luck... meaning what? An error message? Silence?
Presumably it plays to the default output device? But on Windows there
will be some sort of audio mixer that allows you to choose which sounds
from which programs are actually audible.
Just because an error sound, or a beep works doesn't mean that sound
from every other program will automatically be audible.
Hmm. If I try that I get:
FAIL sox: Sorry, there is no default audio device configured
and - depite trying this & that, and googling I've not been able to make
it work. A few posts suggest that it might work on the earlier version
of sox, v14-4-1. I've not tried. Running all sorts of commands, with
-V4 (verbose output) I just see things like:
sox.exe DBUG sox: Looking for a default device: trying format
`waveaudio'
sox.exe DBUG waveaudio: waveOutOpen(QUERY: Dev -1 0Hz 0Ch 8Prec 8Wide)
returned 32
sox.exe FAIL sox: Sorry, there is no default audio device configured
I'm just a user; I have no idea how to fix this. Other programs on my
machine happily produce sound.
Generating a synthesised tone and putting in a file (and later playing
that by double-clicking the .wav file) does work. I'm not a fan of the
'play' & 'rec' command forms though and prefer to build up a standard
sox
command so that all the parameters used are those I specified. Because
of the way a command has global options then for each file that's named
file-options followed by the file name/id, then effect names and options
my view is that one needs to know which is which. So for example in
play -n -c1 synth sin %-12 sin %-9 sin %-5 sin %-2 fade h 0.1 1 0.1
the -n is the input file, which means there's no 'global options' or
first file options... because if there had been they'd have had to
come before "-n".
Then we have "-c1" (which means "one channel") and refers (because it
is a file format option) to the following file. But there's no file
explicitly named after that "-c1". But it's there because "play"
implies an output file. Then "synth .... fade..." are effects and
their parameters.
I prefer to use eg
sox -n -c1 "%temp%\mono1.wav" synth sin %-12 sin %-9 sin %-5 sin %-2
fade h 0.1 1 0.1
or (for much more info as it runs)
sox -V4 -n -c1 "%temp%\mono2.wav" synth sin %-12 sin %-9 sin %-5 sin
%-2 fade h 0.1 1 0.1
Both those generate a file in your Temporary files folder.
[Normally I use scripts to generate sox commands, and I consider global
options,
each file's options etc separately in the script so that I can keep
track of
which part is which, then finally build the command out of each section.
That
way I can make sure that certain options, eg "--no-clobber", are
specified
every time. That one is useful as it prevents accidental overwriting of
any
output file, if one gets the overall command wrong... which is easy to
do.
--
Jeremy Nicoll - my opinions are my own
_______________________________________________
Sox-users mailing list
Sox-users@xxxxxxxxxxxxxxxxxxxxx
https://lists.sourceforge.net/lists/listinfo/sox-users
