In the Android SDK the PeerConnectionFactory allows setting a custom AudioDeviceModule, which then adds the possibility to feed custom audio through calls, instead of the default device microphone.
However, on iOS this seems impossible.
I can tackle this issue and submit a PR myself, but would like to confirm whether it's intentionally (and if so, why) not exposed to library users?