8.8 KiB
alsa
ALSA is part of the kernel.
What is PCM or a PCM device?
https://en.wikipedia.org/wiki/Pulse-code_modulation
aplay
aplay is part of the package alsa-utils on Fedora.
aplay -l
List all soundcards and digital audio devices.
aplay -L
-L, --list-pcms
List all PCMs defined
(there is also default)
You can use aplay to play audio files.
Without argument it works on wav/raw files:
aplay example.wav
With -f there are different formats available, but you would need to be an audio expert to understand what these mean.
speaker-test is another useful binary from alsa-utils:
speaker-test -l 1
This will produce some noise on the default playback device for 10 seconds
and print some info about it on the screen.
See man speaker-test for more.
Also useful:
amixer
amixer info
You can also specify the device
amixer -D default
(and you can also specify the card with -c)
amixer controls
There is also the package alsa-topology-utils which provides the command alsatplg.
configuration
$ cat /etc/alsa/alsactl.conf
#
# ALSA library configuration file for alsactl tool
# (/usr/share/alsa tree is not necessary for alsactl)
#
ctl.hw {
@args [ CARD ]
@args.CARD {
type string
default "0"
}
type hw
card $CARD
}
And there is stuff here:
/etc/alsa/conf.d $ ls
50-pipewire.conf 99-pipewire-default.conf
select default soundcard and/or pcm device
https://www.alsa-project.org/wiki/Setting_the_default_device
=> that works
Find your desired card with:
cat /proc/asound/cards
and then create /etc/asound.conf with following:
defaults.pcm.card 1
defaults.ctl.card 1
Replace "1" with number of your card determined above.
you can test with speaker-test -l 1 or with the minimal pcm example code
Note: The problem is, that your USB headset might be card 0 one time and when you replug it another time it might be card 1.
audioservers
pulseaudio and pipewire are audio servers.
They are on top of alsa.
how to figure out if you are using pulseaudio or pipewire?
$ ps -ef | grep -i "pipe"
bf 2079 1856 0 19:09 ? 00:00:10 /usr/bin/pipewire
bf 2082 1856 0 19:09 ? 00:00:13 /usr/bin/pipewire-pulse
pipewire doesnt seem to have a systemd unit
But there seems to a compatibility layer involved since pulseaudio volume control (or pavucontrol) does change my audio output volume.
pipewire
/usr/share/pipewire/pipewire.conf
The program qpwgraph is supposed to be a gui help for pipewire.
Seems very chaotic though.
pipewire - record audio
If you use pipewire you can test your input device by recording like this:
pw-cat -r recordme
pipe audio to pipewire
cat Hitman_Budapest_Bath_Hotel.mp3 | paplay
pulseaudio
installation
sudo pacman -S pulseaudio
sudo pacman -S pavucontrol
sudo reboot
pulseaudio commands
pactl info
paplay <audiofile>
testing pulseaudio
pactl load-module module-sine frequency=440
stop the sound with
pactl unload-module module-sine
btw, to see all loaded modules:
pactl list modules
ffmpeg
Puts the audio of a .mp4 Video into a .mp3 file:
ffmpeg -i input.mp4 output.mp3
Converting to .webm works the same way:
ffmpeg -i input.mp4 output.webm
Reducing the resolution of a video:
ffmpeg -i <input> -vf scale=480x360,setdar=4:3 <output>
play sounds on the command line
Make sure "system sounds" are almost as high as the other sounds in pulse audio volume control.
mpg123 -q /path/to/myfile.mp3
or directly with pulseaudio:
paplay /usr/share/sounds/freedesktop/stereo/alarm-clock-elapsed.oga
test microphone
part of the libpulse package:
parecord file.mp3
pacman -F parecord
extra/libpulse 16.1-6 [installed]
usr/bin/parecord
usr/share/bash-completion/completions/parecord
The pulseaudio-utils package has the parec command:
parec --format=s16le --channels=2 --rate=44100 | pacat
This will record from your microphone and output it a couple of seconds later.
Volume problem in movies
In movies the volume is always inappropriate, shootings and music is very loud, quiet conversations you want to hear are unhearable silent. In my experience while using headphones this problem is less dominant but still exists.
Using OpenShot: Imported movie, separate audio command took forever so I quit.
Tried it another time. Waited 2 minutes switched to browser in the meantime. Then it worked. But how do I just cut the audio without cutting the video clip too? => don't use short cut, use the mouse, left click to select the clip, right click on the clip -> slice -> keep both sides Then select the clip, right click and put the volume to 10% or delete the audio clip.
=> this works ok, but the performance is a little slow, maybe reduce the quality of the video before doing this!
Exporting was a problem at 720p 30fps medium quality. Took 1 hour and the file is now 4.3GB big which is way too much. The problem was the original file was 720X320 at 24 fps and the new file is 720x1980 at 30fps. So I chose the wrong export option.
Rendering/exporting video has to be done in advanced mode. You need to view the video and audio settings of the original video in file browser (video bitrate, audio bitrate, sample rate, codec, dimensions). The file size of the exported video is ok now, but I still have problems with the resolution. The video is squished together and the black frames on bottom and top are bigger than orignal even though I selected the same resolution. Außerdem Audiotrack verschoben...
VLC Media Player:
Tools -> Effects and Filters (ctrl + E)
=> try experimenting with enabling equalizer and/or enabling compressor ==> doesn't really seem to help, maybe you have to change some settings ===> changed settings, doesn't help
Try that didn't get me anywhere:
sudo apt install pulseaudio-equalizer
command to start the program: qpaeq
Error solution: https://linuxhint.com/install-pulseaudio-equalizer-linux-mint/
sudo vim /etc/pulse/default.pa
Add the below-given lines at the bottom of the/etc/pulse/default.pa file. load-module module-equalizer-sink load-module module-dbus-protocol
Next, kill the PulseAudio service and restart it: $ pulseaudio --kill && pulseaudio --start
Now, open the PulseAudio with the command: $ qpaeq
=> Ok, jetzt funktioniert zwar das Programm, aber ich kann mit dem Programm nichts anfangen. Scheint auch keine Dateien zu manipulieren sondern nur generell den Output.
audacity
sudo apt install audacity
Cut out audio: Select a range with the mouse and hit del
Keep everything but the selected: select a range with the mouse and go to edit -> remove special -> trim audio. (or ctrl + t
Then put it on the beginning of the track with tracks -> align tracks -> start to zero
location of sound files that applications use
chatgpt:
The location of sound files used by a Linux application can vary depending on the application and the specific Linux distribution being used. However, there are a few common locations where sound files are typically stored:
-
/usr/share/sounds: This directory contains system sounds that are used by many Linux applications, such as notifications and alerts.
-
/usr/share/sounds/{application-name}: Some applications may have their own subdirectory within /usr/share/sounds where they store their sound files.
-
~/.local/share/sounds: This directory contains sound files for user-specific applications, such as media players or games.
-
/usr/lib/x86_64-linux-gnu/pulseaudio: This directory contains system sounds for the PulseAudio sound server, which is used by many Linux distributions.
-
/usr/lib/x86_64-linux-gnu/gstreamer-1.0: This directory contains sound files used by GStreamer, a popular multimedia framework used by many Linux applications.
For example this file existed both on Arch Linux and on Fedora:
$ paplay /usr/share/sounds/freedesktop/stereo/alarm-clock-elapsed.oga
troubleshooting
no audio input from headset
I had no audio input from my headset.
The solution was to go into pulse audio volume control or pavucontrol.
"Input devices" -> And set your headset as fallback
With pipewire you can test it with pw-cat -r myfile, go into nnn and hit enter on the file.
no sound issue on Ubuntu
I had the issue that no sound was working on Ubuntu anymore because of various system problems.
Then I did
sudo apt reinstall alsa-base
sudo apt reinstall pulseaudio
reboot
And sound was working again.