Wow, crickets.

Let's see if I may be of some assistance. To answer the question, SciPy or PyAudio are two Python packages that may be used for noise reduction.

To stray off topic to noise reduction, this is not something trivial, since different noise has different properties. This is bad news, because the kind of noise you may encounter in the audio files differs between samples, so it's going to be difficult to have a one size fits all denoiser. I think that's partly why the different packages tend not to have a magic denoiser that just removes noise and boom you're done.

There are a few different approaches I have seen in Python:

1. Build a noise profile, and try to denoise by removing that noise profile from your samples. This is the general approach from most noise reduction attempts I have encountered
2. Copy an existing denoising algorithm, such as the one used by Audacity
3. Spectral gating, which is the approach used by Audacity
4. Use filters, for example high pass or low pass filters

Hope this helps!