[CELT-dev] PEAQ(CELT)

Jonathan M. Polom s0nic0nslaught at gmail.com
Wed Jun 1 06:40:26 PDT 2011


Osama:

Looking at the documentation for the PQevalAudio function, those two
last arguments allow you to specify at exactly what frames the PEAQ
analysis begins and stops. They are both two element arrays because
you can set a different start or end point for the reference and
tested audio. My guess is this is most useful when your reference
audio is perhaps something like a studio recording but the audio under
test was a recording of a transmission channel and isn't necessarily
the same length.

Taken from the PQevalAudio.m file source:
> function PQevalAudio (Fref, Ftest, StartS, EndS)
> % Perceptual evaluation of audio quality.
>
> % - StartS shifts the frames, so that the first frame starts at that sample.
> %   This is a two element array, one element for each input file. If StartS is
> %   a scalar, it applies to both files.
> % - EndS marks the end of data. The processing stops with the last frame that
> %   contains that sample. This is a two element array, one element for each
> %   input file.  If EndS is as scalar, it applies to both files.

I'd definitely have to say though that this topic isn't necessarily
for this list and would be best addressed by the authors of the PEAQ
implementation you're using. They may be able to provide much more
insight.

On Sat, May 28, 2011 at 8:41 AM, Osama Fadhil <osamaesl at yahoo.com> wrote:
>
> Yes, here it is
>
> http://www-mmsp.ece.mcgill.ca/documents/Downloads/PQevalAudio/
>
> After you run the MATLAB, look for an m-file called PQevalAudio.m in the software of the PEAQ
>
> You can type the following in the command line of matlab:
>
> >> PQevalAudio (' Name of the reference (original) file ', ' Name of the tested (de-coded) file ',  parameter here ( scalar or two element array) , parameter here (scalar or two element array) )
>
> Note that this PEAQ software tests files sampled at 48 kHz
>
> My question was how to use the third and the fourth parameters. How should I select them? Depending on what?
>
> Regards,
>
> Osama
> From: Jonathan M. Polom <s0nic0nslaught at gmail.com>
> To: celt-dev at xiph.org
> Sent: Friday, May 27, 2011 6:07 PM
> Subject: Re: [CELT-dev] PEAQ(CELT)
>
> Osama: I'm similarly interested in evaluating CELT with an automated/objective audio and speech quality analysis method. Do you have a link to the PEAQ implementation you mentioned? I might be able to help. Also, could you post the code your analysis uses?
>
> Thanks,
> Jon
>
> On Wed, May 18, 2011 at 8:55 AM, Osama Fadhil <osamaesl at yahoo.com> wrote:
>
> Hello,
>
> I don’t know whether this is the right place to discuss such a question here. I’m sorry.
>
> I want to evaluate the CELT codec using a known MATLAB implementation of the PEAQ software. If anyone has used this test program he could see that one of the input parameters (3rd one) shifts the frames for both the reference and the under test audio signals. Another parameter (4th one) marks the end of data. My question is how to choose such parameters? Depending on what? Is there any operation that we can do on the reference and the tested audio signals in order to know the amount by which we shift the frames?
>
> Thank you for any idea
>
>
> Osama.
> _______________________________________________
> celt-dev mailing list
> celt-dev at xiph.org
> http://lists.xiph.org/mailman/listinfo/celt-dev
>
>
>
>
> --
> Jon Polom
>
>
> _______________________________________________
> celt-dev mailing list
> celt-dev at xiph.org
> http://lists.xiph.org/mailman/listinfo/celt-dev
>
>



--
Jon Polom



More information about the celt-dev mailing list