Skip to main content
Conference ProceedingsOPEN ACCESS

Evaluation of lineal telation between shifted delta cepstral features and prosodic features in speaker verification

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (2008) 5197 LNCS 112-119
DOI: 10.1007/978-3-540-85920-8_14
1Citations
Citations of this article
2Readers
Mendeley users who have this article in their library.

This article is free to access.

Abstract

Abstract. This paper examines the linear relation between Shifted Delta Cepstral (SDC) features and the dynamic of prosodic features. SDC features were reported to produce superior performance to Δ features in Language Identification and Speaker Recognition. A selection of more correlated SDC features is used in speaker verification to evaluate its robustness to channel/handset mismatch. The experiment reflects superior performance of selected SDC features regarding to Δ features in speaker verification using speech samples from NIST 2001 Ahumada database. © 2008 Springer-Verlag Berlin Heidelberg.

Author supplied keywords

Cite

CITATION STYLE

APA

Calvo, J. R., Ribas, D., Fernández, R., & Hernández, G. (2008). Evaluation of lineal telation between shifted delta cepstral features and prosodic features in speaker verification. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 5197 LNCS, pp. 112–119). https://doi.org/10.1007/978-3-540-85920-8_14

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free