Looking at a few loopback apps, capturing the audio stream from a specific application seems to be possible. Would it be a lot of effort to get this feature in Vuo as well? Without an external audio interface, capturing/loopback interfaces are by far the biggest pita when it comes to implementing audio reactiveness in the compositions. I usually hook up an external device when performing with the compositions, but for programming I’m just too lazy to set it up/don’t have it at home (it is also difficult enough balancing the laptop on my lap when travelling by boat).
Would Soundflower work for your purposes?
Soundflower is unfortunately too buggy. I get a lot of latency and dropouts/scratching/noise in combination with Vuo, and it’s a bit of fumbling about every time (not much, but still…).It is also grabbing the audio at the main outs.
I’m thinking more about something like rogue amoebas solution Audio Hijack (Rogue Amoeba | Audio Hijack: Record Any Audio on MacOS). This lets you grab the audio coming at an application level instead of grabbing the i/o. I think this is a feature within the core audio api as far as I’ve understood.
I could maybe look into it myself, but this would be a bit (a lot) on the challenging side for me. Also not sure where the I/O interfacing with the OS would recide in the Vuo Api.
OK, good to know. We’ll have to do some research before we can open this for voting. Not sure how a macOS application is supposed to capture per-application audio these days. It used to be that you implemented a HAL driver, but that seems to no longer be supported. If you happen to find any info about this (or decide you want to take it on yourself), please let me know. I imagine the structure of the code for this would look something like the
vuo.audio.receive node class,
VuoAudio library, and other related libraries.
Super! I tried again today bypassing soundflowerbed which I’ve read could be a bit wonky and/or deprecated, but it was even worse (latest download from Matt Ingalls).
Well, we still need to do some research, but I’ll go ahead and open this for voting.
It would be nice if this feature allowed you to direct the audio to a specific output channel as well. That way you could direct say a microphone to one channel, and music to another, and have them separately drive different attributes; or just for keeping the channels split on a recording.