Whenever I have a conference call the audio comes in as opus@48000 and then gets converter down to slin@8000 for mixing just to be converted back up to opus@48000. My concern with this is the audio quality is much lower when this happens in conference calls than during direct calls when opus@48000 is sent directly.
I am wondering if there is any place to change it so that it mixes at opus/slin@48000 instead of degrading to slin@8000? I have tried changing “internal_sample_rate” to 48000 in confbridge.conf but have seen no change. I have gone through all the sample config files but can’t find a reference to the default conference format.
It dynamically adjusts based on what channels are in the conference bridge. What is the specific assortment of channels and such?
I’ve got one channel coming from a device and another channel from software listening to the device in the conference.
They are both on PJSIP, the device being listened to has ReadTranscode opus@48000>slin@48000>slin@8000 and the listening software has the WriteTranscode opus@48000>slin@48000>slin@8000.
I am expecting that these would change with the mixing rate adjustment. Is there somewhere I should be monitoring this?
Enabling core debug and looking at the output of ConfBridge may show what is going on.
I don’t see much output. Since I am doing a one-way call is it possible to simply broadcast the audio as opus instead without degrading to slinear in a conference?
Are you using Originate and local channels?
Both channels are coming through PJSIP, I am not certain where to check if they are using Originate or local channels.
I can confirm now that no local channels are being used
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.