RNNoise is a noise suppression library based on a recurrent neural network.
RNNoise-nu is a modified version of RNNoise aimed to be a bit more general
purpose.

To compile, just type:
% autoreconf -i
% ./configure
% make

Optionally:
% make install

While it is meant to be used as a library, a simple command-line tool is
provided as an example. It operates on RAW 16-bit (machine endian) mono
PCM files sampled at 48 kHz. It can be used as:

./examples/rnnoise_demo <number of channels> <maximum attenuation> [model to use] < input.raw > output.raw

The output is also a 16-bit raw PCM file.

For information on the neural network models available, see
https://github.com/GregorR/rnnoise-models