.. _parametrised:

Parametrised model compensation
--------------------------------

*Model compensation* is a technique for noise-robustness.
It takes a model of the speech and a model of the noise, and produces
a model of the corrupted speech.
In this toolkit, the models of the speech and noise must both be Gaussian.
Only static speech recogniser coefficients are compensated.

The cross-entropy toolkit implements the following well-known model compensation
techniques.

..  autofunction:: noise.dpmc.dpmcCompensate

..  autofunction:: noise.vts.vtsCompensate