[Speex-dev] About Acoustic Echo Canceller

[Ken Smith]

Sorry nobody responded on this earlier, but for some reason it just came
into my inbox this morning.

Briefly, no, you don't have to convert the samples to the frequency domain
before submitting them to the echo canceller.  The Speex AEC will
automatically convert them to the frequency domain as part of its
processing, and will then convert them back after they've been cancelled.
 If you look at the testecho.c file, you can see how it should be done: it's
assuming two inputs, one a raw (time domain) signal sent to the speakers,
the other a raw (time domain) signal received from the microphone.

Ken Smith
Cell: 425-443-2359
Email: ken at alanta.com
Blog: http://blog.wouldbetheologian.com/


On Thu, Nov 4, 2010 at 10:12 AM, Alexandre Bento Leal <msbleal at gmail.com>wrote:

> Hello.
> I'm from Federal University of Rio Grande do Sul - Brazil and I'm trying to
> adapt speex_echo.h (speex-1.2beta3-win32) to our video-conference software
> code.
> We are working on amplitude values, but I realised that in order to AEC
> works, one has to convert amplitude to frequency and filter the echo
> frequencies. So, my question is:
>
> Do I have to manually convert amplitude samples to frequency samples or is
> it enough to send amplitude parameters to speex_echo.h functions that this
> functions automatically convert it and do the AEC?
>
> Thanks for your attention.
>
> --
> Alexandre Bento Leal
>
> Graduando em Engenharia de Computação - UFRGS
> Membro PRAV - Projetos em Áudio e Vídeo
>
>
> _______________________________________________
> Speex-dev mailing list
> Speex-dev at xiph.org
> http://lists.xiph.org/mailman/listinfo/speex-dev
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.xiph.org/pipermail/speex-dev/attachments/20101112/dba87d55/attachment-0001.htm