Hey all,

I am making a patch based on random playback of audio files [readsf~] selected by external folder. I want to take it a bit further, avoiding the complete randomness, so I am searching for ideas to tag these external files according to their basic frequency, harmonics or amplitude using probably [sigmund~], then store these values into a table / histogram and recall them for playback using specific criteria (i.e play files between certain frequencies or amplitudes). My basic question is how I will save / tag each external playbacked file with these specific properties in order to recall it later in the patch.

happy new year!

@tokeno You mention storing values into a table.
[text] could store the file name and the attributes that you discover using [sigmund~]... and then you can recall the filename by searching lines within [text] for your attributes.

Then you mention changing the filename to store the attributes within the name.

Both are viable, but you don't need both.

However, there is a danger with both that when you ask for a set of attributes they do not exist in your table or in any file name.
So [text] might be best, as you can fall back to a "closest match"..... using < <= near etc. arguments within [text search].
David.

@whale-av thanks! after a long day searching I have arrived into the same conclusion using [list] & [text]. I am now able to save the filename as text with [text set] & [text define]. Now I need to combine the saved filename with its attributes I am getting from [sigmund~].

@tokeno said:

I am searching for ideas to tag these external files according to their basic frequency, harmonics or amplitude using probably [sigmund~], then store these values into a table / histogram and recall them for playback using specific criteria

This sounds like a job for Fluid Corpus Manipulation: https://flucoma.org/

hjh

@ddw_music said:

https://flucoma.org/

In a very very smaller scale, more or less this is a part of what I am trying to achieve. Basically, analyzing field recording and re-combine them according to their properties for the generation of new algorithmic compositions. Amazing link! Will look into it and download asap.