You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Packetize always sets the Marker bit for the last payload in a sequence. I don't believe that that is semantically appropriate. The marker bit is defined to be application- (codec-)specific.
For applications which send either no packets or occasional comfort-noise packets during silence, the first packet of a talkspurt, that is, the first packet after a silence period during which packets have not been transmitted contiguously, SHOULD be distinguished by setting the marker bit in the RTP data header to one. The marker bit in all other packets is zero. The beginning of a talkspurt MAY be used to adjust the playout delay to reflect changing network delays. Applications without silence suppression MUST set the marker bit to zero.
Packetize always sets the Marker bit for the last payload in a sequence. I don't believe that that is semantically appropriate. The marker bit is defined to be application- (codec-)specific.
For example, for voice codecs, RFC 3551 section 4.1 says:
If you feed a sequence of pre-constructed media samples in, as in https://github.com/Sean-Der/audio-spew/blob/687cf84003383d5a9ced1384264d9b99f669cfe5/main.go#L88-L100, every single packet gets the Marker bit set, which is definitely not right.
I suspect that the correct thing for this package to do is to never set the Marker bit, leaving it to the parent application to set it as needed.
cc @Sean-Der
The text was updated successfully, but these errors were encountered: