[asterisk-dev] Zaptel Echo Cancellation

Steve Underwood steveu at coppice.org
Thu Aug 3 17:10:37 MST 2006


Koopmann, Jan-Peter wrote:

>On Thursday, August 03, 2006 6:50 PM Hans Petter Selasky wrote:
>
>  
>
>>Hardware echo cancellers perform no better than software echo
>>cancellers. 
>>    
>>
>
>In theory that is correct of course (at least to my opinion). But if you compare the results some people seem to have with let's say Sangoma cards and hardware echo cancellation compared to all their efforts with Asterisk software echo cancellers the statement does not appear to be applicable. :-)
>  
>
You are comparing a well designed echo canceller, with something very 
crude which addresses none of the key issues with relible cancellation. 
This has nothing to do with software versus hardware.

>My echo went away after tweaking txgain, rxgain and changing to MG2 with echocancel=256 (32ms). Just switching back to KB1 produced immediate problems like ECFO and echo. 128 taps (16ms) also was not enough for my environment. I would not say hardware ec is better or worse than software ec. It has a lot to do with the quality of the algorithm, does it not?
>  
>
Steve





More information about the asterisk-dev mailing list