[BUG] Does not use specified audio device (Always uses `default`)

[second2050]

Desktop (please complete the following information):

• OS: Arch Linux
• KDE Plasma Version 5.21.5
• KDE Frameworks Version 5.82.0
• Using pipewire-pulse instead of pulseaudio

Describe the bug While using the PulseAudio backend the plasmoid only uses the default audio device regardless of what's set in the config window.

Plasmoid Viewer Log panon.log

[rbn42]

How did you know it only uses the default audio device? Is it shown in pgrep -af panon

[second2050]

Okay my bad, I said default audio device because it seemed like it only used the default.

Output of pgrep if set to: "Monitor of Current device":

14936 python3 -m panon.backend.client ws://127.0.0.1:34729 --backend=soundcard --device-index=smart --fps=75 --reduce-bass --bass-resolution-level=4 --enable-spectrum-data

"Monitor of EMU20K2 [SoundBlaster X-Fi Titanium Series]…":

15515 python3 -m panon.backend.client ws://127.0.0.1:34729 --backend=soundcard --device-index=alsa_output.pci-0000_05_00.0.analog-stereo.monitor --fps=75 --reduce-bass --bass-resolution-level=4 --enable-spectrum-data

"Monitor of Family 17h (Models 00h-0fh) HD Audio Controller…":

15606 python3 -m panon.backend.client ws://127.0.0.1:34729 --backend=soundcard --device-index=alsa_output.pci-0000_0b_00.3.analog-stereo.monitor --fps=75 --reduce-bass --bass-resolution-level=4 --enable-spectrum-data

---

In all cases it shows the input of my microphone instead of the what's playing on these devices though.

[rbn42]

Please install python-soundcard, and try this standalone python script. I want to know could this script record audio data from your other devices, or only from your microphone.

import soundcard as sc
_id="alsa_output.pci-0000_0b_00.3.analog-stereo.monitor"
mic=sc.get_microphone(_id)
stream = mic.recorder(44100,2,44100)
stream.__enter__()
while True:
  data=stream.record(44100)
  if sum(data)==0:
    print('muted',sum(data))
  else:
    print('activated',sum(data))

[second2050]

Output of the script:

Traceback (most recent call last):
  File "/home/second2050/rbn42-test.py", line 3, in <module>
    mic=sc.get_microphone(_id)
  File "/usr/lib/python3.9/site-packages/soundcard/pulseaudio.py", line 317, in get_microphone
    return _Microphone(id=_match_soundcard(id, microphones, include_loopback)['id'])
  File "/usr/lib/python3.9/site-packages/soundcard/pulseaudio.py", line 345, in _match_soundcard
    raise IndexError('no soundcard with id {}'.format(id))
IndexError: no soundcard with id alsa_output.pci-0000_0b_00.3.analog-stereo.monitor

[rbn42]

Oh, sorry, I made a mistake at line 3. please try this script again

import soundcard as sc
_id="alsa_output.pci-0000_0b_00.3.analog-stereo.monitor"
mic=sc.get_microphone(_id,exclude_monitors=False)
stream = mic.recorder(44100,2,44100)
stream.__enter__()
while True:
  data=stream.record(44100)
  if sum(data)==0:
    print('muted',sum(data))
  else:
    print('activated',sum(data))

[second2050]

Output of the new script was:

Traceback (most recent call last):
  File "/home/second2050/rbn42-test.py", line 8, in <module>
    if sum(data)==0:
ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

[second2050]

I just did a system update, pipewire-pulse in particular got an update, and it's working again as it should. My guess is that this was a bug introduced by Pipewire and not Panon, therefore I will close this issues for the time being.