Forward audio from BlackHole to the output device.

forwardAudio.py

#!/usr/bin/env python
import sounddevice as sd
import sys

# Download the pkg file of BlackHole
# https://github.com/ExistentialAudio/BlackHole/releases
# Install BlackHole
#
# Install sounddevice
# pip install sounddevice --user
#
# Make file executable
# chmod +x ./forwardSound.py
#
# Click on Mac volume control (on top) and choose 'Output Device': 'BlackHole 16ch'
#
# Check with the following command what name the Built-in Output device has
# python -m sounddevice
sd.default.device = 'BlackHole 16ch', 'Built-in Output'

def callback(indata, outdata, frames, time, status):
    if status:
        print(status)
    outdata[:] = indata

with sd.Stream(channels=2, callback=callback):
    print('#' * 80)
    print('press Return to quit')
    print('#' * 80)
    if sys.version_info[0] < 3:
        raw_input()
    else:
        input()

Hi, thank you for creating this. I'm testing this to see if I can get audio output to my speakers from the BH device like this:
sd.default.device = 'BlackHole 16ch', 'MacBook Pro Speakers'

I am sending output from DjayPro (but also tried Tidal) to the BH device.
but all I get is: "output underflow"

any ideas?

Mmmm.. Does it work when you use 'Built-in Output'?
Sometimes I also see "output underflow". Possibly it even works but is not loud enough?

I've been playing with this a bit, also testing the wire.py example script. I'm sending audio from Tidal/DjayPro into the BH device. My goal would be to get that audio from BH output to the speakers. But still getting no audio at all so far.

Good luck mate. Please let us know if you get it to run.

OK, couldn't get that to run, but was able to use GarageBand to get the BlackHole --> MacBook Pro Speakers working and that's enough for my needs at this time. Thank you!

THANK YOU FOR THIS @PRosenb

@seabass9 and others trying to get it running - couple of things to check:

• Need to have numpy installed, run the command python3 -m pip install numpy. Numpy is kind of a power tool we might not strictly need here, would be cool if somebody changed the code so it could run without numpy. Seems to be supported by sounddevice but I already had numpy so I didn't try it.
• Make sure you are using the same python 3 install both to install packages and run the script. You can be sure by running it like,

       python3 -m pip install numpy sounddevice --user
       python3 ./forwardSound.py```
  

Also, I needed a little lower latency as I was using this to monitor myself playing keys while I stream online (ableton -> blackhole 1/2 for OBS, 3/4 for monitor) so I tweaked a few things that helped in my case: https://github.com/karlthorssen/monitor.py

In the meantime, there is a nice solution available if you like to control the volume of e.g. optical audio output.
For such cases, I recommend Proxy Audio Driver.