Open Bug 1763385 Opened 3 years ago Updated 3 years ago

Virtual device for audio multi-input device integration testing

Categories

(Core :: WebRTC: Audio/Video, task)

All
Unspecified
task

Tracking

()

People

(Reporter: padenot, Assigned: padenot)

References

(Depends on 1 open bug)

Details

We need more than one audio input device, and potentially the capability to have them produce audio.

The preliminary plan is as-is:

  • On macOS, install blackhole-16ch (brew install blackhole-16) to have a second audio input device (there is always a built-in, it seems like). We can then use the other side of the blackhole loopback to produce whatever audio we feel like. For this we'll need to change the macOS image of the test runners
  • On desktop Linux, use PulseAudio to have virtual sources that output sines (so that we have at least one platform where we can determine what device it is based on the audio content). We can also use a loopback device to do thing that are more complicated like on macOS. We can do this without modification of the test runner image.
  • On Windows, it seems like there is already something like VirtualAudioCable, to be confirmed
  • On Android, not sure, and we don't implement it for now, there is generally a single audio input device, and it's the default device, determined by system policy
Blocks: 1238038
Blocks: 1763403
Depends on: 1764901
Depends on: 1412915
You need to log in before you can comment on or make changes to this bug.