[LAU] Getting Rhythmbox to Appear in qjackctl
len at ovenwerks.net
Sat May 23 20:15:14 CEST 2020
On Fri, 22 May 2020, Samir Parikh wrote:
> Thanks for the note. I wanted to give a quick update on my situation. I'm
> not familiar enough with JACK yet to determine whether or not your advice
> still applies. If is still does, I'm not sure I follow everything so may
> have some follow up questions.
> Again, I'm trying to combine the music playing from Rhythmbox with my voice
> from the microphone and send that out to a meeting service like Jitsi (or
> With Rhythmbox running and playing music, I open up qjackctl and the
> "Connections" window looks like this:
Shows your audio device and two pulse-jack bridges, good so far. So Jitsi
now can use jack as it's microphone and speaker/phones. Good.
> In PulseAudio Volume Control, the "Playback" tab looks like:
> The "Output Devices" tab looks like:
That is pretty busy :) I would suggest going to the "Configuration tab"
and setting the "Profile" of any of the devices you don't use to "Off".
Actually I prefer to set them all to off and use jack for all i/o but your
case may be different as you will want things to work right when jack is
not running too. In any case any device you can turn to "Off" is one less
device your applications might choose to deal with.
> With this setup AND with JACK running, I can hear the music through my
> speakers! (Minor victory, but I'll take it!)
> How to I combine my voice from the microphone and then have that serve as the
> "input" to Jitsi or Zoom.
> Is that where these mixers come in or can I do through directly through
> qjackctl and PusleAudio Volume Control?
These mixers are for convenience, you can work without them just fine.
Once you have things running as above, in a terminal on the command line
pactl load-module module-jack-sink client_name=Rhythmbox channels=2 connect=no
The connect=no may not be what you want. If left out, this set of ports
will be connected to system:playback_1/2 which may be what you want so you
can hear rythum box in your speakers. When you have put that command in. A
new entry should show up in your qjackctl Connections dialog. A stereo set
called Rhythmbox:front-left/right should have shown up and may be
connected to the speaker output or not (depending on if you used
connect=no or not). In Rhythmbox setup you should be able to choose this
as an output and if connected to system:playback you should be able to
hear Rhythmbox through your speakers.
Now to use this output into Jitsi, use your qjackctl connections dialog to
connect Rhythmbox:* to PulseAudio Jack source:*
You should now have both your mic and Rhythmbox working as inputs to
Jitsi. So in Pavucontrol, the input level to Jitsi will control _both_ you
Mic level and Rhythmbox level. alsamixer may allow controlling you mic
level or it may be a knob on your audio device. The Rhythmbox level can be
controlled from pavucontrol's Applications tab. If you find that Rhythmbox
sounds "Echo-y" in your speakers, that is a sign that you probably need to
use the connect=no above, otherwise not. Remember that you mic will pick
up whatever is going through your speakers which may be sightly delayed
from you dirrect feed via jack as well as taking room echos of that sound,
so it may be best not to have Rhythmbox attached to your system:playback.
So in case you missed it... Just connecting two sources to a jack port,
mixes the two signals together.
The advantage of creating a second output bridge from pulse to jack over
setting up Rhythmbox to create jack ports is that the pulse bridge ports
are always there and the connections in jack are static and can be created
right after the bridge is instead of waiting for Rhythmbox to create ports
when play is pushed. That stuff will all happen on the Pulse side.
One of the reason's I don't like the pulse bridge's auto creation is the
awkward naming it uses: PulseAudio Jack source, I normally shorten this to
just pulse-in. Normally I would suggest that because you are using ubuntu,
that you just install ubuntustudio-controls which makes setting this stuff
up much easier, but the version of -controls for ubuntu 16.04 does not
have the features you are most likely to want and the newer version only
backports as far as 18.04. However, creating a script for your use case
would not be that hard either.
I have tried to be a clear as possible but may still have made assumptions
about what you already know that are not true. Feel free to ask for
More information about the Linux-audio-user