Kyma Forum
  Tips & Techniques
  random audio slicing

Post New Topic  Post A Reply
profile | register | preferences | faq | search

next newest topic | next oldest topic
Author Topic:   random audio slicing
Tom Heuzenroeder
Member 		posted 01 December 2009 01:47         Edit/Delete Message   Reply w/Quote
Hi everyone.

The following query may seem obvious for most, as I am still getting used to Kyma - I have a lot to learn yet, but someone may have made a Sound for this already:

I am wanting to process incoming audio by dicing it into sections of between say 0.25 of a second to 2 seconds in length, and reassembling the sections in a new/random order - a bit like beat cutting. (see for example: http://mdsp.smartelectronix.com/doc/index.php/Livecut )

I have no doubt that this is easily achievable in Kyma - my purpose for it is not to cut loops/beats, but to process walla/loop-group tracks for film ambiances, to make them "M&E safe" so that any discernible language patterns are destroyed, yet the audio still has the correct flavour for the film.

I have heard of other sound designers using this technique successfully with processors in "Pluggo" - though though Kyna would no doubt give better control over parameters.

I have tried to build the sound by modifying the prototypes and building it that way, but as yet - no luck.

Would anyone be able to offer any guidance?

Thanks in adavnce,
Tom

IP: Logged

SSC
Administrator 		posted 01 December 2009 08:30         Edit/Delete Message   Reply w/Quote
Have you experimented with mixes of SampleClouds with TimeIndexJitter set very high?

IP: Logged

Tom Heuzenroeder
Member 		posted 01 December 2009 17:57         Edit/Delete Message   Reply w/Quote
Hello - thanks for your reply.
Yes I have tried this to an extent, with mixed results. Of course it is very dependent on the audio being processed, but so far I have found that grainclouds make it feel either a bit layered and too thick OR at times the intelligibility is left in tact a bit too obviously.
Having said that though, I reiterate that I am still getting used to it all, and that there is likely a solution in this method that I haven't discovered yet. I'll keep trying, and of course still open to any suggestions. I'll give the grainclouds another go.

Cheers
Tom

IP: Logged

SSC
Administrator 		posted 02 March 2010 21:02         Edit/Delete Message   Reply w/Quote

attemptatm&esafewalla.kym
 
I came across this thread again and thought even if you've solve it already, this kind of thing probably comes up more than once. Attached is a very rough attempt. What do you think sounds right and what sounds off about it? Maybe we can refine it and come up with a useful patch for doing this in general.

IP: Logged

Tom Heuzenroeder
Member 		posted 02 March 2010 21:19         Edit/Delete Message   Reply w/Quote
Ah ok!!! I'll download it right away and try it soon! Thank you for looking at this again. If it can be refined to a few variable parameters which when tweaked can account for most situations, then it will be a most valuable tool, not only for me, but for others wishing to attempt the same thing I'm sure.

Since I first posted, I did resort to other software, but they never really produced completely satisfactory resutls. I still think Kyma will deliver the best. Thanks again - I'll be in touch.

Tom

IP: Logged

SeanFlannery
Member 		posted 07 March 2010 23:12         Edit/Delete Message   Reply w/Quote
Charlie Norton posted a sound in his repligator basics topic called SampleLoopRepligation.kym
that can seriously mash a vocal sample - turns one voice into an instant crowd
http://www.symbolicsound.com/cgi-bin/forumdisplay.cgi?action=displayprivate&number=7&topic=000333

IP: Logged

CharlieNorton
Member 		posted 10 March 2010 07:13         Edit/Delete Message   Reply w/Quote
http://colourofsound.8088.org/This%20Thing%20Used%20To%20Kyma.mp3

Here is some audio I made with the looping repliagations. I know it does get a bit oppressive to listen to after a while. About 2/3rds way though there are some textures which might be something towards what you are after. Obviously the modulations are not fine-tuned for your purpose yet.. Something a bit more bespoke required! (less rhythm more random...)

Charlie

IP: Logged

ChristianSchloesser
Member 		posted 10 March 2010 12:48         Edit/Delete Message   Reply w/Quote
There is an open source software i used before i started with Kyma.
http://www.meapsoft.org/docs.php

i think thats the easyes way to go... it has lots of crazy "extraction" and "sorting" features for chunks...

with kyma i would go the sample cloud way...maybe a sample cloud of a sample cloud etc... ;-) there are so many ways in kyma to do it...

all the best.
chris

IP: Logged

All times are CT (US)

next newest topic | next oldest topic

Administrative Options: Close Topic | Archive/Move | Delete Topic
Post New Topic  Post A Reply

Contact Us | Symbolic Sound Home

This forum is provided solely for the support and edification of the customers of Symbolic Sound Corporation.


Ultimate Bulletin Board 5.45c