Robust speaker verification with principal pitch components

We are presenting a new method that improves the accuracy of text dependent speaker identification systems. The new method exploits a set of novel speech features that is derived from a principal component analysis (PC) of voiced speech segments. The new PC features are only weakly correlated with the corresponding cepstral features. A distance measure that combines both, cepstral and PC pitch features provides a discriminative power that cannot be achieved with cepstral features alone. It is well known that the discriminative power of cepstral features declines if the dimensionality of the feature space is increased beyond its optimal value. By augmenting the feature space of a cepstral baseline system with PC pitch features we are able to reduce the equal error probability of incorrect customer rejection versus incorrect impostor acceptance by 12.5% beyond the discriminative limit of the cepstral analysis.