We consider the problem of parameter estimation in full-covariance Gaussian mixture systems for automatic speech recognition. Due to the high dimensionality of the acoustic feature vector, the standard sample covariance matrix has a high variance and is often poorly-conditioned when the amount of training data is limited. We explain how the use of a shrinkage estimator can solve these problems, and derive a formula for the optimal shrinkage intensity. We present results of experiments on a phone recognition task, showing that the estimator gives a performance improvement over a standard full-covariance system.
Bibliographic reference. Bell, Peter / King, Simon (2008): "A shrinkage estimator for speech recognition with full covariance HMMs", In INTERSPEECH-2008, 910-913.