Unfortunately, what you’re asking for falls outside the scope of support. However, I can definitely point you in the direction of some resources.
For filtering, the “Filtering” chapters of the Digital Signals Theory book by Brian McFee is a good start, and contains Python examples of a number of filter types. You may also want to look into Linkwitz-Riley filters, which will allow you to split a signal into frequency bands - from there, you can process the result and then sum the individual bands together to produce the output.
As for denoising, you can use the new Multiband Dynamics effect included in FMOD 2.03 as a gate with the downward expansion mode. If you’re wanting to write your own denoiser, an easy option would probably be to use a phase vocoder algorithm, which you can use to zero out specific frequencies with magnitudes below a threshold. This is often done using a reference or sidechained “noise” signal.