A New Feature Normalization Scheme Based on Eigenspace for Noisy Speech Recognition

We propose a new feature normalization scheme based on eigenspace, for achieving robust speech recognition. In particular, we employ the Mean and Variance Normalization (MVN) in eigenspace using unique and in– dependent eigenspaces to cepstra, delta and delta-delta cepstra respectively. We also normalize training data in eigenspace and get the model from the normalized training data. In addition, a feature space rotation procedure is introduced to reduce the mismatch of training and test data distribution in noisy condition. As a result, we obtain a substantial recognition improvement over the basic eigenspace normalization.