Start rtc engine prior to someone else joining/ unmuting local microphone

[paulxuca]

Is your feature request related to a problem? Please describe.
I need to access the avaudioengine created by the rtc engine but the rtcengine is only created when one remote participant joins or I unmute my local microphone track. I would like to be able to "force" the rtc engine to be created as soon as I join a Livekit room. I understand the current behavior might be due to reducing bandwidth to the server when not needed but I'm self hosting Livekit so my usecase is slightly different.

Describe the solution you'd like
Be able to "start()" the rtcengine programatically

Describe alternatives you've considered
Creating a fake remote participant in a remote browser to "trick" the sdk into thinking it needs to initialize the rtc engine. Seems really silly and probably won't work.

TIA!

[hiroshihorie]

Hi can I know your use case to access the AVAudioEngine early ?

Currently you need to hook an AudioEngineObserver and configure nodes when requested by the SDK.
It is designed this way since the SDK can create / deallocate the AVAudioEngine multiple times in some cases.

[paulxuca]

It's so I can play audio buffers (sound effects) on the audio engine. Using AVAudioPlayer creates some issue when participants join/ leave since the volume of the playing player changes drastically going from quiet to loud or loud to quiet due to the microphone input node potentially switching on/off.

Thanks for your response!

[istevanovic]

Once another user joins the call. I am unable to play audio queues aswell (for example, when user joins/leaves or when he raises a hand - datachannel message). audio works fine when I am alone on the call.