Advances in objective estimation of perceived speech quality

We present two techniques that can be used to enhance objective estimators of perceived speech quality. Frame normalization and frame-energy plane partitioning are described and applied to a log-spectral-error-based estimator. The resulting estimators are compared with each other and with two established estimators. This is done through correlation with MOS values from 17 formal subjective tests. We find that the proposed techniques significantly improve the log-spectral-error-based estimator.