I have a Asterisk 16 server running MixMonitor for call recordings:
exten => s,n,MixMonitor(${CDR(uniqueid)}.wav,ab,/usr/local/scripts/convert ${CDR(uniqueid)})
The problem I am experiencing is that the call recording is self sounds like it has jitter but the call it self is completely fine.
The covert script does the following to assist with call recording size:
/bin/nice -n 15 /usr/bin/sox /var/spool/asterisk/monitor/$UNIQUEID.wav -g /var/spool/asterisk/monitor/$DATE/$UNIQUEID.WAV
Are the original .wav files OK? It could be the conversion to gsm that is responsible.
We have done some tests and the original wav file seams to be fine.
Do MixMonitor have alternative extensions that will assist with compressing the size of the recording file?
Regards
Ask not what your Asterisk can do for you – ask what you can do for your Asterisk.
Asterisk is already doing a lot of stuff and if you are not careful about the resources, you quickly run into problems. It’s not that it doesn’t work, but the typical computer is not a real time system that operates like a Swiss train: punctual and with guaranteed resources. My approach to your problem was to save a simple file and let a post record command do any necessary conversion and transportation to other servers.
try this
s,n,MixMonitor(${CDR(uniqueid)}.gsm,ab)
or for .wav49 wrapper on GSM data
s,n,MixMonitor(${CDR(uniqueid)}.wav49,ab)
https://wiki.asterisk.org/wiki/display/AST/Asterisk+Audio+and+Video+Capabilities
I have done more testing and used wireshark pcap to look at the rtp traffic:
I seam to have dead spots if the playback timer is set to jitter buffer but with RTP timestamp is looks and sound allot better:
PCAP file is between the provider and the Asterisk 16 server.
I don’t have any jitter buffers programmed. Should I look at programming a jitter buffer on the dial plan?
I think you should be analyzing the wireshark data in more detail, then addressing the network problem.
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.
