This customer Jason is referring to has their audio files setup with no lead-in 
silence (0ms). I’ve seen vague references that 100-500ms is a good best 
practice. Does anyone have any definitive references or experience as to what 
the Best Practice for lead in silence should be?

BTW packet capture (and rtp playout with wireshark) on the CUBE proved the 
audio was being sent to the carrier but is being truncated in the Carrier 
Network.

Thanks,

Joe

From: cisco-voip [mailto:[email protected]] On Behalf Of Jason 
Aarons (AM)
Sent: Thursday, November 06, 2014 2:37 PM
To: cisco-voip ([email protected])
Subject: [cisco-voip] CUBE to VXML the prompt starts playing before call is 
setup


Customer is running UCCE with a VXML gateway that has a CUBE in front of it.

The VXML prompts start to play before the CUBE sets up the audio, so once the 
audio comes in the end user has missed the first 1-2 words.  Normally I tell 
the customer to start record the prompts with a 2 second delay before speaking.

Any ideas why the CUBE/VXML audio isn’t cutting thru fast enough? Debug ccsip 
messages on cube/vxml.  They are using Early Offer as well.


itevomcid
_______________________________________________
cisco-voip mailing list
[email protected]
https://puck.nether.net/mailman/listinfo/cisco-voip

Reply via email to