Hi,
I am looking into the SFUs out there as a customer of mine needs to build a videconferencing sevice based on WebRTC.
Looking at Jitsi and comparing it to Licode I have this question …
Is the Licode SFU supporting and using Audio level extension (RFC6464) to allow a client to decide how many partiipants it wants to see but always get from the server the Active Speaker and the last N speakers?
The use case is the following: Conference with 20 participants:
From a Mobile client I want to receive only the current speaker + 3 most recent speakers (due to screen size and computing bugdet)
On a Desktop I may want to see up to 8 and during the call decide how many I want to see so need to be able to tell the server how many I want to receive.
Additionally if lastN is supported then an Adaptive lastN mechanism is supported too so that the server can send less if the client BW decreases
Thanks