[Csnd] Opcodes, Plugins or UDOs that compute audio descriptors?
| Date | 2019-03-08 20:51 |
| From | Oscar Pablo Di Liscia |
| Subject | [Csnd] Opcodes, Plugins or UDOs that compute audio descriptors? |
Hello Csounders:
Are there Opcodes, Plugins or UDOs that compute audio descriptors from
an given audio file or analysis file?
Many thanks
--
Dr. Oscar Pablo Di Liscia
Profesor Titular
Director Programa de Investigación "Sistemas Temporales y Síntesis
Espacial en el Arte Sonoro"
http://stseas.web.unq.edu.ar/
Director Colección Editorial "Música y Ciencia"
Escuela Universitaria de Artes
Universidad Nacional de Quilmes
Argentina
Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here |
| Date | 2019-03-10 19:51 |
| From | Oeyvind Brandtsegg |
| Subject | Re: [Csnd] Opcodes, Plugins or UDOs that compute audio descriptors? |
Hi Oscar, I did some audio feature analysis for the project "Crossadaptive processing as musical intervention" (http://crossadaptive.hf.ntnu.no). Victor and Rory also contributed to this project. The analysis is wrapped as a Cabbage plugin so it can be used in a DAW, and it sends the feature descriptors over OSC. The csound code is here: https://github.com/Oeyvind/featexmod If you want to use just the audio analysis and put it in your own context, then you can look at the file "analyze_audio.inc" to get the code. You will need to include the file "analyze_udos.inc" too, and also to set initial values for some parameters as you can see in "analyze_chn_init.inc". The whole rhythm analysis part is a mess (sorry!), but the rest of the analysis is good. I also use some filtering of the spectral analysis output (centroid, flux etc) so that they are not so "wild" when the signal is low. For example, the centroid will skyrocket when the amplitude is low, and I apply a smoothed gate to minimize that effect. A few days ago, we also released an album with music from the project: all best Oeyvind Den fre. 8. mar. 2019 kl. 21:50 skrev Oscar Pablo Di Liscia <oscarpablodiliscia@gmail.com>: Hello Csounders: Oeyvind Brandtsegg Professor of Music Technology NTNU 7491 Trondheim Norway Cell: +47 92 203 205 http://www.partikkelaudio.com/ http://crossadaptive.hf.ntnu.no http://gdsp.hf.ntnu.no/ http://soundcloud.com/brandtsegg http://flyndresang.no/ http://soundcloud.com/t-emp |
| Date | 2019-03-10 20:35 |
| From | "Dr. Richard Boulanger" |
| Subject | Re: [Csnd] Opcodes, Plugins or UDOs that compute audio descriptors? |
Hello Oeyvind, WOW. What an amazing project! I am listening to, and loving the ALBUM release right now. - thanks for sharing all the work and developing such an informative webPage about the project. It will be cool to try using some of your Cabbage instruments and Csound tools. Best, -dB PS: Will you be at the ICSC in Italy? Will you be sharing/presenting some of this work there? Hope so. _____________________________________________ Dr. Richard Boulanger Professor of Electronic Production and Design Professional Writing and Music Technology Division Berklee College of Music ______________________________________________ OFFICE: 1126 Boylston St., Suite 201 (EPD), Suite 208 (Boulanger) ______________________________________________ President of Boulanger Labs - http://boulangerlabs.com Author & Editor of The Csound Book - http://mitpress.mit.edu/books/csound-book Author & Editor of The Audio Programming Book - http://mitpress.mit.edu/books/audio-programming-book ______________________________________________ about: http://www.boulangerlabs.com/#about about: http://www.csounds.com/community/developers/dr-richard-boulanger/ music: http://www.csounds.com/community/developers/dr-richard-boulanger/dr-richard-boulanger-music/ ______________________________________________ email: rboulanger@berklee.edu facebook: https://www.facebook.com/richard.boulanger.58 On Sun, Mar 10, 2019 at 3:52 PM Oeyvind Brandtsegg <oyvind.brandtsegg@ntnu.no> wrote:
|
| Date | 2019-03-10 22:06 |
| From | luis jure |
| Subject | Re: [Csnd] Opcodes, Plugins or UDOs that compute audio descriptors? |
i remember that quite some time ago someone asked here about the
possibility of implementing vamp plugins as csound opcodes:
https://www.vamp-plugins.org/
quality and usefulness vary considerably, but most of them have been
developed by experts in the field. having at least some of them
implemented as csound opcodes would be fabulous, if at all possible.
--
Csound mailing list
Csound@listserv.heanet.ie
https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND
Send bugs reports to
https://github.com/csound/csound/issues
Discussions of bugs and features can be posted here |
| Date | 2019-03-11 12:42 |
| From | Oeyvind Brandtsegg |
| Subject | Re: [Csnd] Opcodes, Plugins or UDOs that compute audio descriptors? |
Richard, yes, I am aiming for participation at ICSC, thanks, looking forward to see you again! And Luis, yes, that was me asking about an interface to the Vamp plugins. However, the architecture for Vamp is designed for offline processing and integration with a streaming or realtime implementation is nontrivial it seems. For some types of simple analysis there might be workarounds, but a generic Vamp/Csound interface would be clunky. For this reason I changed my focus to just implementing the types of analysis that I needed directly in Csound code. Some of it was lifted from the "Timbre Toolbox" paper by Peters et al, and Victor implemented the opcodes needed for mfcc analysis. I had an ok method for transient analysis from a previous project, and added rhythmic density analysis to that. There are some measures of rhythmic complexity inspired by ideas from Clarence Barlow, and also measures of consonance from just intonation. I was also looking at Georg Boenns book after the fact, and there is considerable overlap, and also scope for refinement of my methods. Den søn. 10. mar. 2019 kl. 23:06 skrev luis jure <ljc@internet.com.uy>: i remember that quite some time ago someone asked here about the Oeyvind Brandtsegg Professor of Music Technology NTNU 7491 Trondheim Norway Cell: +47 92 203 205 http://www.partikkelaudio.com/ http://crossadaptive.hf.ntnu.no http://gdsp.hf.ntnu.no/ http://soundcloud.com/brandtsegg http://flyndresang.no/ http://soundcloud.com/t-emp |
| Date | 2019-03-11 16:59 |
| From | Oscar Pablo Di Liscia |
| Subject | Re: [Csnd] Opcodes, Plugins or UDOs that compute audio descriptors? |
| Many thanks Oeyvind and Luis, I will take a look at this. Of course i know and use the Sound Visualiser application and its facilities to host plugins of the Vamp norm. And as for the code may be to link to the Essentia library (MTG) could be another straight forward approach. I just was wondering if anyone did something in the field before. All your data are very useful. Best. El lunes, 11 de marzo de 2019, Oeyvind Brandtsegg <oyvind.brandtsegg@ntnu.no> escribió:
-- Dr. Oscar Pablo Di Liscia Profesor Titular Director Programa de Investigación "Sistemas Temporales y Síntesis Espacial en el Arte Sonoro" http://stseas.web.unq.edu.ar/ Director Colección Editorial "Música y Ciencia" Escuela Universitaria de Artes Universidad Nacional de Quilmes Argentina Csound mailing list Csound@listserv.heanet.ie https://listserv.heanet.ie/cgi-bin/wa?A0=CSOUND Send bugs reports to https://github.com/csound/csound/issues Discussions of bugs and features can be posted here |