You didn't try the by far most popular option - OBS Studio.

To get the best audio results:

• add proper source for your input device, e.g. microphone
• add filters to the created source to process the audio on the fly, e.g. remove noise, add some gain and limit max dB treshold. There are lots of videos on e.g. YouTube explaining how to properly set up filters.
• add any other sources for whatever you want to record, specific windows are recommended. If you need other sound from e.g. the system, using Application Audio Capture source is recoomended for the best control.

Framerate applied to both streams, but since ffmpeg documentation examples are scattered I'll leave an answer here

A="$(pacmd list-sources | grep -PB 1 "analog.*monitor>" | head -n 1 | perl -pe 's/.* //g')"
F="$(date --iso-8601=minutes | perl -pe 's/[^0-9]+//g').mkv"
V="$(xdpyinfo | grep dimensions | perl -pe 's/.* ([0-9]+x[0-9]+) .*/$1/g')"
ffmpeg -loglevel error -video_size "$V" -f x11grab -i :0.0 -f pulse -i "$A" -f pulse -i default -filter_complex amerge -ac 2 -preset veryfast "$F"

where

#A=1
#F=2018121711440500.mkv
#V=2560x1440

• ffmpeg the tool
• -loglevel error only print errors
• -video_size "$V" resolution of your screen (or less if you only want a subsection recorded)
• -f x11grab record the screen (screen recordings may not be possible on wayland?)
• -i :0.0 the X11 screen ID, (can also add +x,y for offset)
• -f pulse the audio driver
• -i "$A" the id of the audio stream
• -f pulse the audio driver again (maybe not needed?)
• -i default normally the system microphone
• -filter_complex amerge merge the 2 audio streams
• -ac 2 convert the 4 audio channels to 2
• -preset veryfast go light on video encoding to avoid stuttering
• "$F" the output file

Remember that the parameter order matters, and pavucontrol can re-map audio only while ffmpeg is running.

Just add -ad device_you_want_to_record

You might also want to tweak your audio codec.

This site has good examples: https://verb3k.wordpress.com/2010/01/26/how-to-do-proper-screencasts-on-linux/

I had the same issue, and I will explain how I made it work in my system step by step. I hope this can help someone else.

I will leave the links to some websites that were very useful.

The global idea is to install JACK to manage the audio, and bridge JACK to Pulseaudio to be able to control the audio volume.

Step1. Remove Pulseaudio from the system to get a clean JACK installation:

sudo apt-get remove pulseaudio 
sudo apt-get remove --auto-remove pulseaudio 
sudo apt-get purge pulseaudio 

Step 2. Install JACK:

sudo apt-get install qjackctl jackd2

These are: qjackctl/stable,stable 0.9.1-1 amd64, the user interface for controlling the JACK sound server jackd2/stable,stable 1.9.17~dfsg-1 amd64, JACK Audio Connection Kit (server and example clients) During JACK installation, a windows open asking to Enable realtime process priority. I answered yes.

Step3. Once installed, execute JACK, and a graphical interface will open

Step4. Re-install Pulseaudio and module-jack

sudo apt-get install pulseaudio-module-jack
sudo apt-get install pulseaudio

Step5. Install Cadence software to allow bridging Pulseaudio/ALSA with JACK

sudo apt-get install apt-transport-https gpgv
wget https://launchpad.net/~kxstudio-debian/+archive/kxstudio/+files/kxstudio-repos_10.0.3_all.deb
sudo dpkg -i kxstudio-repos_10.0.3_all.deb
sudo apt install cadence

Step6. Configure the system. Here is where you will need to play around depending on your system audio devices. On JACK, open settings (Setup...), and define your sound card. These are mine:

Execute Cadence: Bridge ALSA/Pulseaudio with JACK

Restart the system Step 7. Configure Simplescreenrecoder to listen to your device. Check that it is getting audio from the system and microphone

Probably this will not work at the first time. Play around with the JACK and Cadence configuration.

Some links that helped me: https://www.thelinuxfaq.com/ubuntu/ubuntu-17-04-zesty-zapus/pulseaudio?type=uninstall https://www.maartenbaert.be/simplescreenrecorder/recording-game-audio/#recording-game-audio-and-microphone-at-the-same-time

If you need something different, or I missed a step, please comment. This is not easy, and having clear steps would help so many people.

Good luck everyone

I've used GTK RecordMyDesktop before and was pleased with the result (but note that I have used only once for really simple stuff).

• [X] Must be gratis
• [X] Must work for Ubuntu 12.04
• [?] Would be cool if it had Unity integration *
• [X] Output of video format is not of matter to me **
• [X] (If possible) low system resource usage (RAM, processor usage and such) ***
• [X] Keyboard short cuts to start and stop recording ****
• [X] Would be great if it had a pause function
• [X] I do not require any editing features, or audio recording (including system sounds)

* What you mean by Unity integration? I remember that it did had an (interactive) tray icon.

** I don't remember exactly which filetype(s) it supports, but it's worth mentioning that I found it to be very limited (it's .ogg only IIRC).

*** I'm not able to say precisely about the resource usage, but I remember that I didn't notice any hiccups.

**** It appears that there is.

This command will capture your full desktop: So use this command, whenever you want to record skype-conversation ( or anything else )

ffmpeg -f x11grab -s wxga -r 25 -i :0.0 -sameq /tmp/out.mpg   

OBS-Studio - Open Broadcaster Software##

In OBS you can record pretty much any part of your screen or audio. You can choose options like window, screen etc for video and mic, speakers etc for audio.

By default these all go into the same channel but under Settings>Output you can choose advanced output mode which also allows for multiple audio channels. So you could choose 2 and set the channels accordingly.

right-click in Audio Mixer and choose 'Advanced Audio Properties'

If your HDD allows, you can try to do it this way:

First write uncompressed file:

ffmpeg -f x11grab -s SZ -r 30 -i :0.0 -qscale 0 -vcodec huffyuv grab.avi

here SZ is your display size (e.g. 1920x1080).

After that you can compress it at any time you want:

ffmpeg -i grab.avi grab.mkv

Of course, you can change compression, select codec and so on.

You can achieve this with peek and key-mon.

Install them:

sudo add-apt-repository ppa:peek-developers/stable
sudo apt update
sudo apt install peek key-mon

Then start them:

key-mon --visible_click
peek

Select the area you want to record.

key-mon shows the key presses and clicks, while peek records the video.

• The GUI is rather unintrusive and allow for easy selection of the area to record.
• Free and open source.
• Can record as WebM.
• Tested on Ubuntu 18.04 as of February 2018.
• No sound it seems.

I would recommend Open Broadcaster Software.

It has support for most Linux distros and it has a set dos option. It does take a while to set it up though to get good quality recordings.

The Ubuntu Screencast Team uses gtk-recordmydesktop.

you can read the solution here and here with more details, BTW to record a good audio you NEED to use jack to link the browser's audio output to vlc

so shortly here :

sudo apt-get install pulseaudio-module-jack qjackctl

now you nee d to run qjackctl and set like in this picture

after that you can test here with this : in this case I have used mp3 and x264 like codecs you can just try to change that if you need.
File will be recorded in tmp folder and to stop recording just press Ctrl+C

cd /tmp ; cvlc --jack-input-auto-connect screen:// --screen-fps=25 --input-slave pulse://jack_in --sout "#transcode{venc=x264,quality:50,scale=1,fps=25,acodec=mp3,ab=128,channels=2, samplerate=44100}:duplicate{dst=std{access=file,mux=mp4,dst=nowardev.mp4}}}"

I managed to get it going with the steps on the Ubuntu Forums, for clarity here is what I did:

1. sudo apt-get install gtk-recordmydesktop pavucontrol
2. Opened the Pulse Audio Volume Control dialog: Applications > Sound & Video > PulseAudio Volume Control
3. Opened gtk-recordmydesktop
4. In gtk-rmd advanced preferences, "Sound" tab, set "Device" to pulse
5. In gtk-rmd start a recording
6. In Volume Control goto the Recording tab and change the recordmydesktop entry to 'Monitor of '

This is what seems to have worked for me.

Here is someone's list of the top four Linux screen recorders and how to install them. I hope this list includes what you are looking for.

VLC media player

• Can capture the screen
• Can capture a webcam
• Can record audio (in conjunction with one video input if desired)

I have already posted a manual on how to do the things mentioned above in another answer: https://softwarerecs.stackexchange.com/a/3311/583

I was going to recommend Kino which is the one I used when I needed to do that, but according to their web page the project is no longer mantained, so you can either use it the way it is or try the other software that is recommended on their web page:

Shotcut, Kdenlive, Flowblade, OpenShot, PiTiVi, LiVES, and LightWorks.

GUI

Audacity

If you want a GUI to do this you can do it using Audacity.

excerpt from wiki

In the drop-down menu on Audacity's Device Toolbar, choose “Wave Out” or “Stereo Mix” as the input source. (The exact name may be different, depending on your computer's sound drivers). On Windows, if you don't have a “Wave Out” or “Stereo Mix” option, or if it won't record, go to the system Control Panel and try to enable this option there. For instructions see: Using the Control Panel on the Wiki.

Streamtuner

You can also install streamtuner to browse radio stations.

# Debian/Ubuntu
$ apt-get install streamtuner

# Fedora/CentOS/RHEL
$ yum install streamtuner

screenshot of streamtuner

commandline

You can install streamripper like so:

# Debian/Ubuntu
$ sudo apt-get install streamripper

# Fedora/CentOS/RHEL
$ sudo yum install streamripper

To download a stream simply do the following:

$ streamripper <url>

This will save each stream into a .mp3 file.

References

• How to record rip streaming net radio in Ubuntu Debian Linux easy steps

Let me repeat: A single program or group of programs A (game) should output sound both to OBS and the headphone, while another single program or group of programs B (voice chat) should only output sound to the headphone, all on the Pulseaudio level. Correct?

Don't use snd-dummy, it works on the ALSA level. Instead, create a "null sink" on the Pulseaudio level:

pacmd load-module module-null-sink sink_name=game_sink sink_properties=device.description=Game-Sink

Use pavucontrol or, if it can do that, the sound configuration of elementary OS to switch all group A sound outputs to that sink. Each sink in Pulseaudio comes with a corresponding "monitor" source (you can see those in the OBS menu you included), so setup OBS to record from "Monitor of Game Sink".

That takes care of recording from group A only, but doesn't output it to the headphones. For that, you need a loopback from the mentioned monitor source to the headphone sink:

pacmd load-module module-loopback source="game_sink.monitor" sink="your-headphone-sink"

You can find out the names of all sinks, including the headphone sink, with

pacmd list-sinks | grep name:

Leave out the angular brackets when using the names as arguments.