41. Granular Synthesis with Pure Data

Andrew R. Brown
andrewrbrown.net.au
Real-time Music and Sound with Pure Data vanilla
A series that explores the basics of electronic music through visual programming.
To keep the grain start time inside the audio data when randomising the start time, add a [%] object between [+] and [next] and pass it number of frames output from [soundfiler].

[@timnormand5396]

I'm loving these tutorials Andrew, thank you for providing them!

[@n.l.j]

Very clear and concise patching, thanks!

[@RobiPell-AmatorialElectr-il6mu]

Just followed your tutorial: I'm started studying granular synth techniques, your video seems one of the best and clearer I met. Very very useful! Thanks

[@Dimitris_Pappas]

Very nice! Thank you.

[@WARDISWARD]

I think granular synthesis is just an evolution of timestretch
, I was using the tabplay~ with a short loop length ( Around 30 ms ) ; by adjusting the loopstart ( which also banged the tabplay ) I could scan through it , make it a bit smoother by converting the looplenght (in samples ) to hert used by a phasor + another tabread~as a windwowing function (so tabread~multiplied by tabplay~ )
Crude but fun

[@casim8842]

I have solved the issue of decorrelating grain duration and grain pitch, so changing pitch does not affect grain duration and vice versa. If you're interested.

[@algomedia]

Great. Link your patch here, I'm sure others would be interested.

[@rimbod1366]

Can ideas from this patch be used to make the chopped samples from the audio cutter upper reamain consistent in pitch/length/speed if given the same metronome as the step sequencer?

[@algomedia]

Pitch shifting and time stretching are not trivial, but there is an example to get you started in Pd's help browser - pd/doc/3.audio.examples/G09.pitchshift.pd

[@rimbod1366]

@@algomedia Thank you! I'll give it a look.

[@BurkhardusX]

In my Pure Data it is impossible to connect a vline~ to tabread4~. It says: can't connect audio signal outlet to nonsignal inlet.