Carsten

I am starting to use Max4Live to develop an M4L effect for an audiovisual performance and worked through tons of tutorials and docs the last days. I want objects in a 3D scene to react to or be controlled by the different properties (origin/pan, frequency, loudness, tempo) of the audio of a live spoken / played performance. The audio is recorded, processed and played out in Ableton live on separate channels. I am a bit lost, where to start such a project.

plugsend~ seems to be the only way to route audio from other channels into the patch but has timing issues from what I found. Is it still with Max 7 and are there alternatives? I read several older posts on the limitations but may be that changed with one of the recent versions?

jit.catch~ I am wondering how to use in stereo (I need to follow the actual panorama of the audio, not the panorama control in Ableton life). jit.catch~ can have several inputs and creates a plane for each so I could use them for the L/R audio input of one Ableton audio channel. But how can I map this data to X choordinates or other parameters by comparing them?

Also how to separate certain frequencies to use for different parts of the animation. Do I need to do this upfront with the audio signal first and route it to separate jit.catch~ or can I do this on the matrix data? I am not sure what the data in the matrix represents and how it is applied.

Any help appreceated especially tutorials, books, videos etc on this topic.

Hi and hello to the forum,
I am starting to use Max4Live to develop an M4L effect for an audiovisual performance and worked through tons of tutorials and docs the last days. I want objects in a 3D scene to react to or be controlled by the different properties (origin/pan, frequency, loudness, tempo) of the audio of a live spoken / played performance. The audio is recorded, processed and played out in Ableton live on separate channels. I am a bit lost, where to start such a project.

plugsend~ seems to be the only way to route audio from other channels into the patch but has timing issues from what I found. Is it still with Max 7 and are there alternatives? I read several older posts on the limitations  but may be that changed with one of the recent versions?

Any help appreceated especially tutorials, books, videos etc on this topic.
Many thanks!
C.


understanding-jit-catch-signal-data-to-matrices-and-others

From what i know, jit.catch just fills a matrix with the waveform data. Unless im wrong, i suspect you would want to filter the audio prior to it hitting the jit.catch.

That said, there may be some sort of matrix processing genius who could work with the data that comes out of jit.catch, but i imagine that would be taking the hard road.

That said, there may be some sort of matrix processing genius  who could work with the data that comes out of jit.catch, but i imagine that would be taking the hard road.


for basic audio signal to a floating point number use snapshot~30

jit.catch 2 would be enough for a stereo file. Just plug Left into inlet 1 and right into inlet 2.

I doubt the results would be what you want though.

With jit.catch I normally throw out 1024 (1 x 1024) values to control a particle system, so unless you want to control "lots" of paramaters it won't be what you need.

Imagine you use a high pass filter on a sound, you reduce its total volume, this plugged into snapshot would also give a float which becomes smaller.

Audio processing to each mono channel like High pass, delay,etc etc.

You can use the slide object afterwards to smooth out the data as it will be jittery.

This data can then be sent to Max msp with udpsend.

I've used this kind of set up many times before, if you just add them to return tracks in live and use send on a track then you have a very easy reusable system which is separate from your audio / midi tracks.

for basic audio signal to a floating point number use snapshot~30
jit.catch 2 would be enough for a stereo file. Just plug Left into inlet 1 and right into inlet 2.
I doubt the results would be what you want though.
With jit.catch I normally throw out 1024 (1 x 1024) values to control a particle system, so unless you want to control "lots" of paramaters it won't be what you need.
Imagine you use a high pass filter on a sound, you reduce its total volume, this plugged into snapshot would also give a float which becomes smaller.
I'd say try the following

stereo signal
Audio processing to each mono channel like High pass, delay,etc etc.
Then to 2 snapshot~30  objects
You can use the slide object afterwards to smooth out the data as it will be jittery.
This data can then be sent to Max msp with udpsend.
I've used this kind of set up many times before, if you just add them to return tracks in live and use send on a track then you have a very easy reusable system which is separate from your audio /  midi tracks.


Many thanks Andro! I´ll look into this and try if this works for me.

Understanding jit.catch~ signal data to matrices and others