25 Sep
2004
25 Sep
'04
3:28 p.m.
Hello,
This time I am encoding from a cassette with a rather low noise level.
And I'd like to do several things automatically, from the command line,
rather than call up Audacity and GWC and do some manual selection.
So, which command line tools(if they are available, of course) will:
- Do lowpass filtering?
- Cut off the silence in the beginning and the end?
- Split a file into songs based on long periods of silence in betweeh
them?
For the latter two points, the silence is not absolute, there is some
noise at about -40 db. The tool I want would take a "noise level" value
in db as a parameter. The REALLY good tool would leave about a second of
"silence" before/after a song intact, because it may contain the
beginning or end that are below the threshold. The PERFECT tool would
also be able to look at a file with just noise and give me that db
value, but it may be too much to ask, so I can find out in Audacity as
well.
So - are these tools available? Or am I out of luck? Of course, I can
use GWC to split a file into songs and Audacity to cut out the silence
manually, but that's quite a lot of mouse-dragging for an operation that
could be automatic.
Yours, Mikhail Ramendik
26 Sep
26 Sep
1:07 p.m.
So, which command line tools(if they are available, of
course) will:
- Do lowpass filtering?
- Cut off the silence in the beginning and the end?
- Split a file into songs based on long periods of silence in betweeh
them?
I don't know if Ecasound can do those, but it's worth checking out..
http://eca.cx/
Tommi
1:43 p.m.
Tommi Sakari Uimonen wrote:
> So, which command line tools(if they are
available, of course) will:
> - Do lowpass filtering?
ecasound has several different lowpass filters. There are also highpass
and bandpass filters as well.
> - Cut off the silence in the beginning and the
end?
ecasound has a threshold gate. it is also possible to cut files by time
if you already know how much silence you need to chop off.
> - Split a file into songs based on long periods of
silence in betweeh
> them?
This isn't directly available, afaik, but could possibly be scripted.
I don't know if Ecasound can do those, but
it's worth checking out..
definitely.
-Eric Rz.
2 Oct
2 Oct
6:32 a.m.
Eric Dantan Rzewnicki wrote:
I asked in the ecasound list - but unfotunately got no answer.
To script this, I would need to use the threshold gate, but in a
different way. It would have to stop processing when the gate
opens/closes, so I can get the time point when that happens.
Can ecasound do this? If so, how can I make it do this?
I like the idea of ecasound scripting. But I could not find this
feature, and without it the thing is useless for me. Now with "find
place where gate works", I'd work wonders ;)
Yours, Mikhail Ramendik
> - Split a
file into songs based on long periods of silence in betweeh
> them?
This isn't directly available, afaik, but could possibly be scripted.
3 Oct
3 Oct
5:03 p.m.
Mikhail Ramendik wrote:
Eric Dantan Rzewnicki wrote:
I asked in the ecasound list - but unfotunately got no answer.
To script this, I would need to use the threshold gate, but in a
different way. It would have to stop processing when the gate
opens/closes, so I can get the time point when that happens.
Can ecasound do this? If so, how can I make it do this?
I like the idea of ecasound scripting. But I could not find this
feature, and without it the thing is useless for me. Now with "find
place where gate works", I'd work wonders ;)
-ge:open-threshold-%, close-thold-%,volume-mode
Threshold gate. Initially gate is closed. It is opened when volume
goes over 'othreshold' percent. After this, if volume drops below
'cthold' percent,gate is closed and won't be opened again. If
'value_mode' is 'rms', average RMS volume is used. Otherwise peak
average is used. When closed, passing audio buffers are trucated to
zero length.
I could be wrong and I haven't used this gate myself, But ....
I think you need to do something like this:
1)process original input file with -ge outputting to song1.wav
If I understand "When closed, passing audio buffers are trucated to zero
length." correctly, then nothing will be written to song1.wav after
volume drops below the threshold. So,
2)get length of song1.wav and save it in a variable, like maybe
position_variable.
3)process original input file with -ge outputting to song2.wav with
-y:<position_variable> . -y sets the starting position.
4) get length of song2.wav and add to position_variable
5) return to 2), increment song<n>.wav and repeat untill finished.
hope that helps,
Eric Rz.
>-
Split a file into songs based on long periods of silence in betweeh
>them?
This isn't directly available, afaik, but could possibly be scripted.
26 Sep
26 Sep
2:39 p.m.
Sox will strip silence. Here's what I do with telephone messages
recorded by vgetty:
sox in.wav out.wav silence 0 1 02.00 1%
On Sat, 25 Sep 2004 21:10:43 +0400, Mikhail Ramendik <mr(a)ramendik.ru> wrote:
Hello,
This time I am encoding from a cassette with a rather low noise level.
And I'd like to do several things automatically, from the command line,
rather than call up Audacity and GWC and do some manual selection.
So, which command line tools(if they are available, of course) will:
- Do lowpass filtering?
- Cut off the silence in the beginning and the end?
- Split a file into songs based on long periods of silence in betweeh
them?
For the latter two points, the silence is not absolute, there is some
noise at about -40 db. The tool I want would take a "noise level" value
in db as a parameter. The REALLY good tool would leave about a second of
"silence" before/after a song intact, because it may contain the
beginning or end that are below the threshold. The PERFECT tool would
also be able to look at a file with just noise and give me that db
value, but it may be too much to ask, so I can find out in Audacity as
well.
So - are these tools available? Or am I out of luck? Of course, I can
use GWC to split a file into songs and Audacity to cut out the silence
manually, but that's quite a lot of mouse-dragging for an operation that
could be automatic.
Yours, Mikhail Ramendik
--
De gustibus non disputandum est.
7306
days inactive
7314
days old
linux-audio-user@lists.linuxaudio.org
5 comments
4 participants
participants (4)
-
Eric Dantan Rzewnicki -
Hans Fugal -
Mikhail Ramendik -
Tommi Sakari Uimonen