bird_furniture
New member
Hi, I'm new to Teensy and working on a project which involves some real-time pitch shifting with the Granular effect in the library. Unfortunately, in the synthesized output, there is a very audible click between processed grains, which at higher speeds creates some undesirable resonance. For this project in particular, I'd prefer to squash these artifacts as much as possible to get a cleaner tone.
One method I tried was to use a waveform generator to modulate amplitude, and act as a sort of envelope for each grain. Then, by using another modulated granulator 180 degrees out of phase, I should be able to cancel out the dip in volume to produce a steady tone. This worked in theory, I'm able to get a steady tone from a normal, unprocessed playback, but once the granulator is introduced the approach falls apart. I think it could work if I were able to somehow have the second granulator begin half a grain after the first, or start mid-grain, but I can't figure out how to do this. Using a normal delay function between instances seems to simply delay the entire void setup, and using a delay audio effect seems to only delay the granulator's output, not the actual processing, and as such the waveforms are still in phase.
Does anyone have any tips on how to get this approach to work, or if there might be other ways to mitigate the granular effect's artifacts? Perhaps there's a more manual approach to granular pitch shifting that would allow me to control this?
I'm using a Teensy 4.0
Appreciate any help! Thanks!
One method I tried was to use a waveform generator to modulate amplitude, and act as a sort of envelope for each grain. Then, by using another modulated granulator 180 degrees out of phase, I should be able to cancel out the dip in volume to produce a steady tone. This worked in theory, I'm able to get a steady tone from a normal, unprocessed playback, but once the granulator is introduced the approach falls apart. I think it could work if I were able to somehow have the second granulator begin half a grain after the first, or start mid-grain, but I can't figure out how to do this. Using a normal delay function between instances seems to simply delay the entire void setup, and using a delay audio effect seems to only delay the granulator's output, not the actual processing, and as such the waveforms are still in phase.
Does anyone have any tips on how to get this approach to work, or if there might be other ways to mitigate the granular effect's artifacts? Perhaps there's a more manual approach to granular pitch shifting that would allow me to control this?
I'm using a Teensy 4.0
Appreciate any help! Thanks!