reading the audio buffer from looper live effect

[tiago_vla]

hi guys.. I hope everything is fine with you guys.. I am working on a granular synthesis/sound processing patch based on munger~ and sakodna granular synthesis patch for MaxMSP.. I would like to add a function to this, which would be in the case of the sakodna patch, importing the buffer from live's looper, into the buffer in the sakodna patch, which would be on separate tracks.. anyone has an idea of how to do such a thing.. I still don't understand the maxforlive api very well.. it's being a bit confusing for me.. so I NEED HELP..

thanks for the attention.. greets..
tiago morgado

[monohusche]

probably not possible as the Live Object Model (LOM) only allows to access the generic parts of Live, such as track, device, clip etc.

It doesn't let you access audio buffers, neither for clips nor for the Looper. You would have to implement a M4L-based Looper using it's own buffer.

[pukunui]

This is much requested and is on our radar for a future version.

Currently the only way to get live clip audio is to record it from an audio input into your device, have a look at BufferShuffler

You can drag and drop audio clips in from Live's file browser as well.

-A

[tiago_vla]

I will take a look @ buffershuffler so, in order to implement this thing of granular synthesis into my device.. maybe estabilishing a limit of time to the buffer~ object and then being constantly recording without overdubing to it, might work for the purpose I have in mind..

the thing of sample dropping is extremely easy.. it's a dropfile with an output to a prepend replace message or object and then an income of it into the audio stuff.. I have already implemented it..

thanks for the help..
tiago morgado

[tiago_vla]

http://www.youtube.com/watch?v=Ctsok5NZmwQ

let's fight for making this thing capable of real time audio manipulation in a maxforlive device

[laurentius]

wouldn't it be possible to have at least the selected clip (in clip view) being granularized in the m4l-patch? actually i know that i am not of any help here for this looper buffer issue.
laurentius

[monohusche]

there is no direct access to the clip's audio content, doesn't matter whether it is currently selected or not. in any case, you have to route the audio into the M4L patch which can only be done with the currently playing clip anyway.

[bdhm]

wouldn't it be possible to have at least the selected clip (in clip view) being granularized in the m4l-patch? actually i know that i am not of any help here for this looper buffer issue.
laurentius

yes, that is a work around I am using: I select a sample in the LIVE-grid [with the LEMUR or the mouse or controller] to read the sample name, to load the sample into a MAX/MSP-buffer.
my next step is to build a M4L-looper to sample into a buffer and to write AIFF-files into a folder.
now I have to build something to read the files from the folder automatically into the LIVE-grid.

but access to LIVE's buffer would be great !

bd

[tiago_vla]

In relation to laurentius, I do not need to granularize a clip.. but something that is being looped in real-time and is in the buffer of a looper.. if i needed to granularize the sound that comes from an audio clip, I would probably use FTM from ircam, or munger~ in maxforlive or otherwise the grain delay from live which would probably be even simpler than that.. the thing is that I need to work with something that is being looped in real time from my instrument.. What I will probably do will be giving a look with more attention to buffer shuffler m4l device.. and learn how it is using real time buffer~ writing, to shuffle the sound with chucker~, and then using this technique to granularize the sound in real time, with the looper signal.. if someone gets advances in relation to the other thing (building a loopstation to granularize its buffers), I will help with it..

[monohusche]

not sure whether buffershuffler is the right thing for you. its purpose is to slice a loop and play the slices in a different order. chucker~ allows 64 steps max which allows for grains of minimum of 31 ms (given a 1 bar loop at 120bpm)....quite long for granular players. It's not comparable to proper granularization a la munger~ or gabor (FTM). There are no overlapping grains, no grain pitch manipulation etc.


There is no difference whether you want to process a Live audio clip or input from your instrument, in any case the audio needs to be "streamed" to your M4L, also for BufferShuffler, there is no way around that. before the loop can be shuffled, it needs to play at least once.

so, in summary: if you want to granularize your audio, go for munger~/gabor or existing max granularizers (not sure whether there is one provided with M4L), if you just want to shuffle the slices for your loop, BufferShuffler could be a good starting point.

hope that helps, and I understood your intention.

nick

[tiago_vla]

My purpose is also to make a personal implementation of granular synthesis, in this particular and specific case, hacking the sakodna model, for using a looper buffer as buffer~, with stuff from microsounds and some personal ideas in relation to granular synthesis for an academic work.. that's why we were discussing the possibility of creating a looper for m4l, or otherwise making a stream of audio data, into the device input for then being reading the thing in real time and writing it to a buffer~ without actually overdubing, and being applying granular synthesis to it. I just had the idea that someone told me here, that I could see the buffershuffler example, to see how can I write to a buffer in real-tiem.. I know I could use munger~ or ftm, which would be actually like piece of cake for such a purpose, but what I want to do is being using the buffer of a looper to make this..