[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [AVT] RE: avt Digest, Vol 28, Issue 21

Hi

> > Stupid question. What if the audio is going one way and there IS NO 
> > RTP packet to piggyback the feedback on going in the reverse 
> > direction? Lots of cases for this, e.g. VAD, MOH, playing 
> voicemail, 
> > etc.
> > 
> > Seems like you need the RTCP machinery anyway.
> 
>  VAD (I assume that you mean DTX, correct me if I am wrong 
> here) the case is that silence descriptor packets are 
> transmitted with regular intervals, for AMR the interval is 
> typically 160ms, this interval is however a requirement from 
> a circuit switched cellular network perspective and for 
> packet switced networks 500ms is sufficient.

AFAIK, there is no such requirements in other networks. VAD/DTX are
somewhat codec dependant. If you use G.729 audio for example, you will
not see silence descriptors generated with regular intervals. If the
input "silence" doesn't change in terms of spectrum or intensity, no new
silence descriptor is generated.
You can decide to send the previous one periodically but it is an
implementation choice, it is not required anywhere.

Aurelien

_______________________________________________
Audio/Video Transport Working Group
avt at ietf.org
https://www1.ietf.org/mailman/listinfo/avt