Fitting a Gaussian mixture model (GMM) to the smoothed speech spectrum allows an alternative set of features to be extracted from the speech signal. These features have been shown to possess information complementary to the standard MFCC parameterisation. This paper further investigates the use of theseGMMfeatures in combination with MFCCs. The extraction and use of a confidence metric to combine GMM features with MFCCs is described. Results using the confidence metric on the WSJ task are presented. Also, GMM features for speech corrupted with additive noise are extracted from data corrupted with coloured additive noise. Techniques for noise robustness and compensation are investigated for GMM features and the performance is examined on the RM task with additive noise. A Nonlinear Observation Model for Removing