Please clarify: do you mean conference build with RoomApi or conference based on MCU mixer?
In the first case, RoomApi actually implements signaling only to transfer test chat messages and to and determine the number of participants. So you can publish RTMP stream to WCS, then play it via WebRTC to every conference participant using stream name. For example, if you publish stream using RTMP URL
Code:
rtmp://wcs:1935/live/user1
you can play it in browser by user1 name.
In the second case, you can either publish RTMP stream with a special suffix pointing room name
Code:
rtmp://wcs:1935/live/user1#room1
and the stream will automatically add to [/ICODE]room1[/ICODE] mixer, or publish RTMP stream without the suffix