Hi,<div><br></div><div>I'm looking for any pointers to help me understand Asterisk rtp.c and the handling of SIP-to-SIP bridging for audio.</div><div><br></div><div>The background: I'm having trouble because one of my termination suppliers has started sending RTP streams with 60msec of audio in each RTP packet.</div>
<div><br></div><div>Our Asterisk boxes are bridging that SIP stream out to another SIP and in the process changing the packetisation from 60msec to 20msec.</div><div><br></div><div>The problem with that being that this results in 3 RTP packets being sent in a "bunch". The network between the server and the phone is resequencing those packets, the SNOM320 phone at the edge is not coping with that at all well and the call sounds bad.</div>
<div><br></div><div>I'm trying to get the network fixed, but this is an obscure point and these days its hard to get your hands on someone at the ISP who can even understand the problem, never mind fix it.</div><div><br>
</div><div>In the meantime the problem will be avoided if I could persuade Asterisk to "follow-the-leaders" and maintain the 60msec packetisation on the way out.</div><div><br></div><div>I did find doc/rtp-packetization.txt and I'm going to experiment with autoframing=yes and the other stuff, but I'd appreciate any other hints or pointers to documentation to understand this better.</div>
<div><br></div><div>Thanks.</div><div>Steve</div><div><br></div>