[Expyezp] sintesis concatenativa

josé manuel berenguer jmbeal at telefonica.net
Tue Apr 19 19:32:09 PDT 2011


Hola, 

no sé si será de utilidad, pero hace unos años escribí una serie de objetos para max que realizaban síntesis por onditas (wavelets) por medio de una implementacion del algoritmo de Daubechies. No trabajan en tiempo real. Solo llenan tablas de onda. Como el código está disponible en los archivos zip, quizá pueda ser reutilizado para escribir objetos en PD o SC que trabajen en tiempo real. Son jmb.dauby, jmb.daubya y jmb.daubyb. Están en mi página web en el link a cosas de Max :

http://www.sonoscop.net/jmb/  --> cosas de max

El link directo a cosas de max creo que es :

http://www.sonoscop.net/jmb/max/index.html

Para más información acerca de Ingrid Daubechies :
http://www.google.es/m/url?client=safari&ei=EkOuTeC5DIKd8wOwno26AQ&hl=es&oe=UTF-8&q=http://en.wikipedia.org/wiki/Daubechies_wavelet&ved=0CBwQFjAB&usg=AFQjCNG3AUkuA2kHl9O3oSTdZgx8_7_taw


Mucha salud a todos


José Manuel


jmberenguer at sonoscop.net
www.sonoscop.net/jmb/
+34669820918

El 20/04/2011, a las 02:16, <joseduarte777 at yahoo.com> escribió:

> 
> Parece que hay gente en la lista de SC que está trabajando el tema de Wavelets ver correo abajo. Avisame si quieres que te envie la conversacion completa son como 36 correos:
> 
> Mencionan un plug-in para wavelet en SC llamado WaveletDaub.sc en sc3-plugins/NCAnalysisUgens
> 
> [Ver abajo]
> 
> 
> rom    yvan volochine <yvan.sc at googlemail.com>
> reply-to    sc-users at lists.bham.ac.uk
> to    sc-users at lists.bham.ac.uk
> date    Mon, Feb 14, 2011 at 3:12 AM
> subject    Re: [sc-users] MIT Wavelet Lectures (Prof. Strang)
> mailing list    sc-users.create.ucsb.edu Filter messages from this mailing list
> mailed-by    lists.bham.ac.uk
>     
> hide details Feb 14
>     
> On 02/14/2011 07:08 AM, Dara Shayda wrote:
> 
>     Hello everyone
> 
>     I am studying wavelets and using them in Mathematica 8.0 for potential
>     Auditory Scene Construction i.e. auditory compositions made from objects
>     that allow for zooming in/out and other related functionality that takes
>     us beyond the antiquated FFT and filters:
> 
> 
> for a start there is WaveletDaub.sc in sc3-plugins/NCAnalysisUgens
> 
> cheers,
> _y
> - Show quoted text -
> 
> 
> _______________________________________________
> sc-users mailing list
> 
> info (subscription, etc.): http://www.beast.bham.ac.uk/research/sc_mailing_lists.shtml
> archive: http://www.listarc.bham.ac.uk/marchives/sc-users/
> search: http://www.listarc.bham.ac.uk/lists/sc-users/search/
> Reply
>         
> Forward
>         
> Invite yvan volochine to chat
>     
> yvan volochine    
> for a start there is WaveletDaub.sc in sc3-plugins/NCAnalysisUgens cheers,
>     
> Feb 14
> Reply
> |
> Dara Shayda
>  to sc-users
>     
> show details Feb 14
>     
>     from    Dara Shayda <dara at untiredwithloving.org>
> reply-to    sc-users at lists.bham.ac.uk
> to    sc-users at lists.bham.ac.uk
> date    Mon, Feb 14, 2011 at 3:20 AM
> subject    Re: [sc-users] MIT Wavelet Lectures (Prof. Strang)
> mailing list    sc-users.create.ucsb.edu Filter messages from this mailing list
> mailed-by    lists.bham.ac.uk
>     
> hide details Feb 14
>     
> > so that it should be possible to 
> > write a set of UGens based on a library to actually use this stuff...
> 
> I agree, I looked at the server code extensively and (to lesser degree at the) Bison grammar and I see no limitation to SC to develop wavelets.
> 
> Actually for super large auditory landscapes e.g. thousands of wavelets, SC is most scalable as I discovered experimenting earlier.  
> 
> However my goal is not just to use them for the sake of modernity, I was thinking to use as auditory objects as opposed to signals. Signals in SC are quite limiting the range of the applications, but if it is broken into short-lived wavelets it is much more suitable to build new apps. 
> 
> I suspect that you can load a set of wavelets that comprises an auditory scene, and zoom in and out naturally without much difficulty. Then this gives a new tool for composers:
> 
> http://www.wolfram.com/mathematica/new-in-8/wavelet-analysis/visualize-scalograms-in-three-dimensions.html  
> 
> You cannot manipulate the traditional def of signals, and it is much harder to build some  sort of landscape with them.
> 
> I am sure there have much smarter folk who figured things on earlier, so I am on the hunt. 
> 
> D
> - Hide quoted text -
> 
> 
> 
> On 2011-02-14, at 3:58 AM, nescivi wrote:
> 
> > On Monday 14 February 2011 07:08:00 Dara Shayda wrote:
> >> Hello everyone
> >>
> >> I am studying wavelets and using them in Mathematica 8.0 for potential
> >> Auditory Scene Construction i.e. auditory compositions made from objects
> >> that allow for zooming in/out and other related functionality that takes
> >> us beyond the antiquated FFT and filters:
> >
> > Wavelets have been quite interesting to me for quite a while... I even did a
> > course on it in university, though it was quite theoretical at that time
> > (about 10, 11 years ago...), and I'm sure the theory and applications have
> > developed quite a bit further since.
> >
> > It seems that by now there are some reference implementations around (quick
> > google; haven't checked how useful they are), so that it should be possible to
> > write a set of UGens based on a library to actually use this stuff...
> >
> > Now to find the time...
> >
> > sincerely,
> > Marije
> >
> > _______________________________________________
> > sc-users mailing list
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> 
> ________________________________________________________________________
> 
> www.proyectosonorum.org
> http://www.myspace.com/sonorumcr
> 
> 
> --- On Tue, 4/19/11, Juan I Reyes <juanig at Maginvent.ORG> wrote:
> 
> From: Juan I Reyes <juanig at Maginvent.ORG>
> Subject: [Expyezp] sintesis concatenativa
> To: "expyezp-slow" <expyezp at lists.slow.tk>
> Date: Tuesday, April 19, 2011, 9:45 AM
> 
> 
> Hola,
> 
> Será que alguno de ustedes a ensayado este tipo de síntesis
> concatenativa [1]. Habrá implementación en Pd  o SC?.
> 
> Por lo que entiendo esta basada en Wavelets como Gabor en lugar de la
> FFT.
> 
>   --* Juan
> 
> [1] http://imtr.ircam.fr/imtr/CataRT
> 
> _______________________________________________
> 
> Expyezp mailing list
> Expyezp at lists.slow.tk
> http://lists.slow.tk/listinfo.cgi/expyezp-slow.tk
> 
> Archivo de mensajes:
> http://lists.slow.tk/pipermail/expyezp-slow.tk/
> _______________________________________________
> 
> Expyezp mailing list
> Expyezp at lists.slow.tk
> http://lists.slow.tk/listinfo.cgi/expyezp-slow.tk
> 
> Archivo de mensajes:
> http://lists.slow.tk/pipermail/expyezp-slow.tk/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.randomlab.net/pipermail/expyezp-randomlab.net/attachments/20110420/9dc8157b/attachment-0003.htm>


More information about the Expyezp mailing list