I am setting up to live stream music. I've installed a virtual sound card (Blackhole) and created an aggregate device consisting of my mic input and the virtual output. I then set OBS to use the virtual device as the audio source, and switched on monitoring. It works! I can hear the music and my voice. The problem is that OBS is creating latency on the monitoring, so my voice is out of sync.

So, can I create an aggregate/multi device which has one input, and two outputs? That way I can have the input for the mic, output directly to my sound card so I can monitor, and have the virtual output going to OBS for the live stream. Is this possible? I can't find a way to add an input to a multi-output device, or to add more than one output to an aggregate device.

1 Answer 1


Ah, I actually found a workaround. My music software (Reason) allows you to wire up the audio output to any channel on the device. Because I created an aggregate with the input and the two outputs, I could split the audio and wire it to the output channels for both the real sound card (for monitoring) and the virtual one (for capture by OBS). It works, and there's no latency.

  • Interesting. It would be nice if we could figure out this type of routing without need for a full-blown DAW, for people doing 'amateur' tasks like this. We get a lot of 'How do I route this to that? I've got Soundflower' type questions here & often the answer so far has been Rogue Amoeba, or an external DAC of some kind.
    – Tetsujin
    Commented Mar 24, 2020 at 12:06
  • yeah it's only because I'm a musician that I have access to that kind of software. I would prefer not to have to do it this way as it means reconfiguring every project in the DAW, but I want to live stream this tonight so it's a workaround that isn't too arduous and does the job. I could probably have done it using OBS on a separate computer, but that would be even more work. Thanks for the reply man, good to know it's not just me!
    – mashers
    Commented Mar 24, 2020 at 12:11
  • Indeed. My own audio structure is based around ASIO usually, with more consumer-oriented machines using various Rogue Amoeba structures or hybrids [my cinema Mac runs a bizarre hybrid of SoundSource with Waves plugins ;) I dumped Soundflower many years ago, so when we get this type of question it's difficult to try quick patching to see what may work on a regular consumer-level machine, without tearing my own structure apart too much, so it's nice to know we have someone else here who can test alternative structures perhaps.
    – Tetsujin
    Commented Mar 24, 2020 at 12:26

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .