I'd like to create a new node that will operate much like FFT, that is it should get a buffer of data (say 32ms or 10ms worth of data) from the microphone node and then output a buffer filled with mutated version of the input data.
How can this be done with AudioKit? Can new DSP nodes that operate on fixed length audio frames be created in Swift alone or will I need to implement a "AudioKit/ExtendingCore" node in C++?