LDA Combination of Pitch and MFCC Features in Speaker Recognition

This paper assesses two popular speaker features pitch and cepstral coefficient (static and dynamic). We compare the performance of individual features and features combined via LDA. The identification process can be performed both in the temporal and cepstral domains. The temporal analysis determines which phonemes or utterances exhibit the highest degree of speaker specificity, while the cepstral analysis examines individual cepstra within these temporal divisions.