[asterisk-dev] Zaptel echo canceler as runtime option

[Kevin P. Fleming]

Tzafrir Cohen wrote:

> Basically: will there be a major performance hit if we replace the echo
> cancellers with an array of function pointers or something similar? Even
> being able to set the echo canceller via a module parameter would be a
> step in that direction.

I wouldn't think that would be a concern, no.

> (And the obvious observation is that non of the current echo canceller
> modules works well enough for a large enough subset of the cases,
> right?)

MG2 seems to be working quite well for a lot of people.