[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Measuring audio quality
If it's specifically "quality" you're interested in, and have some listeners handy, you could consider methodologies along the lines of those proposed in ITU-R BS.1534 or ITU-R BS.1116. In this suggestion, or Emmanuel's, you could use the input target signal as the reference.
Chris
http://www.itu.int/rec/R-REC-BS.1116-1-199710-I/e
http://www.itu.int/rec/R-REC-BS.1534-1-200301-I/e
http://en.wikipedia.org/wiki/MUSHRA
On 13 Jun 2011, at 11:50, XAVIER ANGUERA MIRO wrote:
> Thank you Emmanuel for your fast answer. I have looked at the links you
> propose, but I have a doubt whether the tools expect to have a clean
> signal as reference?
> In my tests I obtain several output signals that I would like to compare i
> order to know which one is "better", but I do not have any ground-truth or
> reference.
>
> Thank you,
>
> Xavier Anguera
>
> ------
> Xavier Anguera, Ph.D.
> Research Scientist
> Multimedia research group
> Telefonica R&D
> Tlf: +34 93 365 3203
> Linkedin: http://es.linkedin.com/in/xanguera
> Twitter: http://twitter.com/xanguera
> Website: http://www.xavieranguera.com/
>
>
>
>
>
>
> On 13/06/11 11:18, "Emmanuel Vincent" <emmanuel.vincent@xxxxxxxx> wrote:
>
>> Dear Xavier,
>>
>> Source enhancement or, more generally, source separation algorithms may
>> be evaluated via the energy ratio metrics in
>> http://sisec2008.wiki.irisa.fr/tiki-index.php?page=Under-determined+speech
>> +and+music+mixtures
>> (scroll down the page) or the perceptual metrics in
>> http://bass-db.gforge.inria.fr/peass/
>>
>> Best,
>>
>> Emmanuel
>>
>> --
>> Emmanuel Vincent
>> METISS Project-Team
>> INRIA Rennes - Bretagne Atlantique
>> Campus de Beaulieu, 35042 Rennes cedex, France
>> Phone: +332 9984 2269 - Fax: +332 9984 7171
>> Web: http://www.irisa.fr/metiss/members/evincent/
>>
>>
>> On 06/13/2011 10:44 AM, XAVIER ANGUERA MIRO wrote:
>>> Dear all,
>>>
>>> I am working on an acoustic beamforming system that, depending on the
>>> input parameters, will output a single audio signal. I am not very
>>> familiar with the signal quality area and would like to ask you for
>>> advice as to how to measure the quality of each of the signals. Note
>>> that I have already tried some toolkits available online that give me
>>> some SNR-related measure, but I do not find these consistent across
>>> several signals in my dataset. Maybe some sort of comparison between my
>>> output signals would give me better results?
>>>
>>> Thanks in advance for your advice
>>>
>>> Yours,
>>>
>>> Xavier Anguera
>>>
>>>
>>> ------------------------------------------------------------------------
>>> Este mensaje se dirige exclusivamente a su destinatario. Puede consultar
>>> nuestra política de envío y recepción de correo electrónico en el enlace
>>> situado más abajo.
>>> This message is intended exclusively for its addressee. We only send and
>>> receive email on the basis of the terms set out at.
>>> http://www.tid.es/ES/PAGINAS/disclaimer.aspx
>
>
> Este mensaje se dirige exclusivamente a su destinatario. Puede consultar nuestra política de envío y recepción de correo electrónico en el enlace situado más abajo.
> This message is intended exclusively for its addressee. We only send and receive email on the basis of the terms set out at.
> http://www.tid.es/ES/PAGINAS/disclaimer.aspx