On 14/12/09 02:41, Andreas Engel wrote:> is it somehow possible to use the internal data of the echo canceller to
> get some kind of a double talk detector? This would allow to perform
> echo suppression in scenarios where the cancellation fails, like with
> webcams, where AEC doesn't work at all.
That's already (sort of) how things works. Now, it still doesn't help
you much with the webcam case because if the AEC fails to converge,
it'll also fail to give you any sort of reliable estimate for suppression.
Cheers,
Jean-Marc