7 Nov
2016
7 Nov
'16
11:38 a.m.
Hey hey again,
the next thing I'm looking for is a program to detect volume over a certain
threshold, which can print out a marker there. This function should - at best
- be able to ignore new peaks within a certain period of time, like a gate
with very long release times.
To elaborate the scenario: I'm recording myself sleeping, to a) find
involuntary actions and b) do quick narrations of dreams, should I wake up and
remember anything. So, I'd like to jump to the short moments of "action"
within the relative silence.
Any help is appreciated, thanks for staying with me.
Best wishes,
Jeanette
--------
When you need someone, you just turn around and I will be there <3
7 Nov
7 Nov
12:12 p.m.
This is not exactly what you describe but it may meet your requirements:
http://plugin.org.uk/timemachine/
It's a GUI program, but the GUI is just one big button, which may or may
not have a keyboard binding.
The program has an auto-record feature based on the signal level. here's
the command line help:
Usage timemachine: [-h] [-i] [-c channels] [-n jack-name]
[-t buffer-length] [-p file prefix] [-f format]
[-a] [-b begin-threshold] [-e end-threshold] [-T end-time]
[port-name ...]
-h show this help
-i interactive mode (console instead of X11) also enabled
if DISPLAY is unset
-c specify number of recording channels
-n specify the JACK name timemachine will use
-t specify the pre-recording buffer length
-p specify the saved file prefix, may include path
-s use safer characters in filename (windows compatibility)
-f specify the saved file format
-a enable automatic sound-triggered recording
-b specify threshold above which automatic recording will begin
-e specify threshold below which automatic recording will end
-T specify silence length before automatic recording ends
HTH, Chris
12:31 p.m.
Nov 7 2016, Christopher Arndt has written:
...
...
Hi Chris,
that did look interesting, but I can't compile it, possibly the source
code is a little deprecated. Still thanks a lot, it would have done.
Best wishes,
Jeanette
--------
When you need someone, you just turn around and I will be there <3
2:03 p.m.
On Mon, 7 Nov 2016 13:13:17 +0100, Christopher Arndt wrote:
Am 07.11.2016 um 12:31 schrieb J. C.:
I still have a PKGBUILD available.
You could change the version and release, if required fix the
dependencies and simply replace the checksums with
"sha256sums=('SKIP')", but seemingly 0.3.3 is still the latest version.
Maybe libsndfile is part of the dependency chain, if not, who cares,
if you anyway have it installed.
[rocketmouse@archlinux ~]$ cat /var/aur3/timemachine/PKGBUILD
# Contributor: DonVla <donvla(a)users.sourceforge.net>
pkgname=timemachine
pkgver=0.3.3
pkgrel=3
pkgdesc="A quick one-click audio file recorder for JACK audio applications"
arch=('i686' 'x86_64')
url="http://plugin.org.uk/timemachine/"
license=('GPL')
depends=('jack' 'gtk2')
source=(http://plugin.org.uk/timemachine/${pkgname}-${pkgver}.tar.gz)
md5sums=('6f26d1c4bac96b36dfe743bcc07bfe53')
sha256sums=('3e798a952b83874e3d60642ca4a7e0c2e2990c4659561009087da95044b21f9d')
build() {
cd ${srcdir}/${pkgname}-${pkgver}
LDFLAGS="-lm" CFLAGS="$SLKCFLAGS" CXXFLAGS="$SLKCFLAGS"
./configure --prefix=/usr
make
}
package() {
cd ${srcdir}/${pkgname}-${pkgver}
make DESTDIR=${pkgdir} install
}
Regards,
Ralf
that did look interesting, but I can't
compile it, possibly the
source code is a little deprecated. Still thanks a lot, it would
have done.
On Debian/Ubuntu it's in the package archives as "timemachine".
2:21 p.m.
> I can't compile it
When using the posted PKGBUILD, the compiling works on
[rocketmouse@archlinux tmp]$ uname -m
x86_64
and launching it works, too, but I didn't test if it does its job.
[rocketmouse@archlinux tmp]$ date;sudo pacman -Syu
Mon Nov 7 14:09:53 CET 2016
[rocketmouse@archlinux tmp]$ sed -i "s/pkgrel=3/pkgrel=4/" PKGBUILD
[rocketmouse@archlinux tmp]$ makepkg -s
[rocketmouse@archlinux tmp]$ sudo pacman -U
/var/cache/aur/timemachine-0.3.3-4-x86_64.pkg.tar.xz
[rocketmouse@archlinux tmp]$ grep timemachine /var/log/pacman.log
[2014-07-22 18:56] [PACMAN] Running 'pacman -S timemachine'
[2014-07-22 18:58] [PACMAN] Running 'pacman --color auto -U
/tmp/yaourt-tmp-rocketmouse/PKGDEST.wEV/timemachine-0.3.3-3-x86_64.pkg.tar.xz'
[2014-07-22 18:58] [PACMAN] installed timemachine (0.3.3-3)
[2016-11-07 14:15] [PACMAN] Running 'pacman -U
timemachine-0.3.3-4-x86_64.pkg.tar.xz'
[2016-11-07 14:16] [PACMAN] Running 'pacman -U
timemachine-0.3.3-4-x86_64.pkg.tar.xz'
[2016-11-07 14:16] [PACMAN] Running 'pacman -U
/var/cache/aur/timemachine-0.3.3-4-x86_64.pkg.tar.xz'
[2016-11-07 14:17] [ALPM] upgraded timemachine (0.3.3-3 -> 0.3.3-4)
[rocketmouse@archlinux tmp]$ timemachine
Cannot connect to server socket err = No such file or directory
Cannot connect to server request channel
jackdmp 1.9.11
Copyright 2001-2005 Paul Davis and others.
Copyright 2004-2015 Grame.
jackdmp comes with ABSOLUTELY NO WARRANTY
This is free software, and you are welcome to redistribute it
under certain conditions; see the file COPYING for details
no message buffer overruns
no message buffer overruns
no message buffer overruns
JACK server starting in realtime mode with priority 10
self-connect-mode is "Don't restrict self connect requests"
audio_reservation_init
Acquire audio card Audio0
creating alsa driver ... hw:0|hw:0|256|2|48000|0|0|nomon|swmeter|-|32bit
configuring for 48000Hz, period = 256 frames (5.3 ms), buffer = 2 periods
ALSA: final selected sample format for capture: 32bit integer little-endian
ALSA: use 64 periods for capture
ALSA: final selected sample format for playback: 32bit integer little-endian
ALSA: use 64 periods for playback
Attempting to auto-start LASH server
Not sending deprecated LASH_Client_Name event
Listening for OSC requests on osc.udp://localhost:7133
JackTemporaryException : now quits...
Jack main caught signal 2
Released audio card Audio0
audio_reservation_finish
timemachine: exiting
2:34 p.m.
FWIW jack is jack2 and it's not from official repositories, but I guess
this doesn't make a difference, if you want to build timemmachine.
[rocketmouse@archlinux ~]$ pacman -Q jack2
jack2 1.9.10.r177.g7bdad49-1
3:24 p.m.
Hello,
There is also a nice Qt app also, QJackRcd which can record stereo
through Jack, and automagically pause or create a new file when the
audio reaches a certain threshold. This is useful for very long sessions
with moments of blank audio.
This is included in Debian, Ubuntu and Arch, but if you prefer build it
from scratch, the project is on sourceforge:
https://sourceforge.net/projects/qjackrcd/
Happy recording..
- Benoît
Le 07/11/2016 à 12:12, Christopher Arndt a écrit :
This is not exactly what you describe but it may meet
your requirements:
http://plugin.org.uk/timemachine/
It's a GUI program, but the GUI is just one big button, which may or may
not have a keyboard binding.
1:13 p.m.
"J. C." <julien(a)mail.upb.de> wrote:
Hey hey again,
the next thing I'm looking for is a program to detect volume over a certain
threshold, which can print out a marker there. This function should - at
best - be able to ignore new peaks within a certain period of time, like a
gate with very long release times.
This should be doable with dtas-partstats, part of my very
incomplete and largely undocumented duct-tape audio suite:
http://dtas.80x24.org/
Basically, dtas-partstats runs the sox "stats" effect on 10s
intervals of a longer audio file and outputs the stats for each
interval to stdout. In the output, the 1st column is the time
interval, 2nd column is the combined peak of all channels.
It should not be difficult to code up something similar using
the "trim" and "stats" effects of sox in your favorite scripting
language if you don't have Ruby installed.
To elaborate the scenario: I'm recording myself
sleeping, to a) find
involuntary actions and b) do quick narrations of dreams, should I wake up
and remember anything. So, I'd like to jump to the short moments of
"action"
within the relative silence.
Anyways, I use partstats to find peaks in live recordings for
testing limiter, compressor and EQ effects to even out the
overall volume of the result. For live concerts, these peaks
are are common from audience noise, dropped/bumped equipment,
or room gain on drums.
It can also take advantage of multiple CPUs to process the 10s
partitions in parallel on systems with GNU nproc(1) or Ruby 2.2+
installed,
1:33 p.m.
* J. C. <julien(a)mail.upb.de> [2016-11-07 11:38]:
Hey hey again,
the next thing I'm looking for is a program to detect volume over a certain
threshold, which can print out a marker there. This function should - at
best - be able to ignore new peaks within a certain period of time, like a
gate with very long release times.
To elaborate the scenario: I'm recording myself sleeping, to a) find
involuntary actions and b) do quick narrations of dreams, should I wake up
and remember anything. So, I'd like to jump to the short moments of
"action"
within the relative silence.
I tried to record myself sleeping in order to find out
what I was
uttering. In my case I recorded the entire night, an approach which you
should do as well, and then strip silent parts from the recording in an
extra step. This way you will not lose material in case your filter is
not configured optimally.
I must say that I was more in need of a snooring detector than any other
detectors though. ;)
SoX has two effects, silence and vad (voice activity detector) which can
possibly anaylze your recording.
One would have to put this inside a script that chops off segments of
your recording, and then repeats the same action with the remainder.
A research using ixquick gave the following as well:
https://github.com/amsehili/auditok
The aubio toolkit on Debian offers two binaries:
aubioquiet - a command line tool to extracts quiet and loud regions from a file
and
aubiocut - a command line tool to slice sound files at onset or beat timestamps
which sound promising as well.
1:41 p.m.
Nov 7 2016, Peter P. has written:
...
I tried to record myself sleeping in order to find out
what I was
uttering. In my case I recorded the entire night, an approach which you
should do as well, and then strip silent parts from the recording in an
extra step.
Hey hey Peter,
that already is my approach, on the same grounds that you mention.
...
SoX has two effects, silence and vad (voice activity
detector) which can
possibly anaylze your recording.
Ah, I expected that SoX would be thrown in at some
point. Well, I'll
look into Sox and the other tools you mentioned. Thank you.
...
https://github.com/amsehili/auditok
The aubio toolkit on Debian offers two binaries:
aubioquiet - a command line tool to extracts quiet and loud regions from a file
and
aubiocut - a command line tool to slice sound files at onset or beat timestamps
...
Best wishes,
Jeanette
--------
When you need someone, you just turn around and I will be there <3
7:56 p.m.
Hi Jeanette, List,
I've done this with sox and a bash script.
The script records from default audio device only when there is a sound
to record (i.e. exceeding a THRESHOLD). Each file is encoded as flac (in
order to not consume much disk space for very long runs) and the file
name contains some info such as a timestamp, an RMS value, ans some
other statistics.
I wrote the script, but some parts are taken from varius guide I've
found on the net, so I've not the credits for the original author(s).
Here is the script
#!/bin/bash
FORMAT=".flac"
DEVICE="default"
THRESHOLD="5%"
TMPFILE=tmp$FORMAT
echo $TMPFILE
while true; do
sox -t alsa $DEVICE $TMPFILE silence 1 0.1 5% 1 1.0 5%
DATE=$(date +%Y%m%d%H%M%S)
LEN=$(sox $TMPFILE -n stat 2>&1 | sed -n 's#^Length
(seconds):[^0-9]*\([0-9.]*\)$#\1#p')
MAX=$(sox $TMPFILE -n stat 2>&1 | sed -n 's#^Maximum
amplitude:[^0-9]*\([0-9.]*\)$#\1#p')
RMS=$(sox $TMPFILE -n stat 2>&1 | sed -n 's#^RMS
amplitude:[^0-9]*\([0-9.]*\)$#\1#p')
FREQ=$(sox $TMPFILE -n stat 2>&1 | sed -n 's#^Rough
frequency:[^0-9]*\([0-9.]*\)$#\1#p')
mv $TMPFILE $DATE$FORMAT
echo "[$DATE] $LEN, $MAX, $RMS, $FREQ" >> record1.log
done
On 07/11/2016 13:42, J. C. wrote:
Nov 7 2016, Peter P. has written:
...
--
a.
I tried to record myself sleeping in order to
find out what I was
uttering. In my case I recorded the entire night, an approach which you
should do as well, and then strip silent parts from the recording in an
extra step.
Hey hey Peter,
that already is my approach, on the same grounds that you mention.
...
SoX has two effects, silence and vad (voice
activity detector) which can
possibly anaylze your recording.
Ah, I expected that SoX would be thrown in at
some point. Well, I'll
look into Sox and the other tools you mentioned. Thank you.
...
https://github.com/amsehili/auditok
The aubio toolkit on Debian offers two binaries:
aubioquiet - a command line tool to extracts quiet and loud regions
from a file
and
aubiocut - a command line tool to slice sound files at onset or beat
timestamps
...
Best wishes,
Jeanette
--------
When you need someone, you just turn around and I will be there <3
_______________________________________________
Linux-audio-user mailing list
Linux-audio-user(a)lists.linuxaudio.org
http://lists.linuxaudio.org/listinfo/linux-audio-user
2939
days inactive
2939
days old
linux-audio-user@lists.linuxaudio.org
12 comments
7 participants
participants (7)
-
Alessio Degani -
Benoît Rouits -
Christopher Arndt -
Eric Wong -
J. C. -
Peter P. -
Ralf Mardorf