Forwarding n loudest participant audio from video-bridge

Hi guys…! was out of jitsi for quite some time…

Anyway, as far as I know, Jitsi Video-bridge currently forwards n loudest participant’s audio packet to everyone so that it can support scalability. Does this mean just the top n audio packets with the highest RMS / ENERGY (which has issues with mid-sentence-cut and false switch) value or there is a complex algorithm like DominantSpeakerIdentification?
If so, then can you provide me with the corresponding RFC related to that algorithm Jitsi is using and the files that implemented it in the jitsi project?

Thanks in advance for any kind of help :heart:

Just wanted to check whether I was right or wrong about the loudest n ? is there something I am missing?