since teaching in Greece is currently online via Zoom, we have many
students that use SuperCollider on Windows. While booting scsynth
on MacOS with zoom connected and sharing computer audio is no problem,
on Windows all of the users experience a freezing of audio when
Is there a way to share computer audio on zoom and boot scsynth on Windows?
Note: I have seen a similar topic by @MarcinP and will respond there to
put students in touch for testing his audio device selection work.
I’m using SC + Zoom from a Mac (with an old 10.10), but these points might be worth a try on Windows as well.
.) start SC before Zoom
.) select 48 kHz for the mic in your system settings
.) select the Zoom device from SC (check if this is the right name on Windows)
s.options.outDevice = “ZoomAudioD”;
.) start Zoom
I avoid using any other audio software at the same time, even before the session. It happened twice that I opened an audio editor with the result that SC + Zoom didn’t work anymore before a restart of the computer.
Thanks very much @dkmayer for these instructions.
One of the students who was able to present with sound via zoom from Windows, said that he was using a similar procedure. I will post the instructions to my classes and test this friday in class if they work out for us, also.
I haven’t tested on Windows, but since zoom audio device only works as an output, what might help is to disable inputs in SC (Server.default.options.numInputBusChannels_(0) I think ). Of course this is only practical if one doesn’t need inputs in SC…
VoiceMeeter is a free/donation-ware Windows app that mixes mic and app audio into one stream, which can then feed into meeting software’s audio input. I’ve just successfully used this with SC and Tencent Meeting (Chinese app, since I’m in the mainland). I have no reason to believe it wouldn’t work for Zoom.
In VoiceMeeter, set the desired mic input and audio output devices.
In the meeting software, set audio input to be VoiceMeeter’s virtual output device (this is the mixdown, mic + apps).
In SC, set s.options.outDevice to VoiceMeeter’s virtual input – I used MME, maybe other protocols would work too. I set inDevice to the hardware mic input, to avoid any risk of software feedback.
Boot server. At this point, the soundcard driver complained about an unsupported sample rate (?) but after that, it was fine.
I did have to try this a couple times before getting everything right, but this did hold up well for a whole 80-minute lesson.
(Note that this depends on the ability to choose different input and output devices, which seems fully stable in 3.11.2 thanks to some brilliant work by Marcin IIRC. This may not work with earlier versions of SC in Windows.)