Title :
Data sampling approaches for GMM supervector based speaker verification
Author :
Dikici, Erinç ; Saraçlar, Murat
Author_Institution :
Elektrik ve Elektron. Muhendisligi Bolumu, Bogazici Univ., Istanbul, Turkey
Abstract :
GMM supervectors are among the most popular feature sets used in SVM-based text-independent speaker verification. Most of the studies represent speaker characteristics obtained from a long recording with a single supervector in the SVM space. Working on the NIST SRE´10 dataset, this study compares the effect of two sampling methods to increase the number of supervectors, on the verification performance. Dominance of positive and negative classes on model construction is investigated.
Keywords :
Gaussian processes; speaker recognition; support vector machines; GMM supervector based speaker verification; SVM-based text-independent speaker verification; data sampling approaches; Adaptation model; Conferences; NIST; Signal processing; Speaker recognition; Speech; Support vector machines;
Conference_Titel :
Signal Processing and Communications Applications (SIU), 2011 IEEE 19th Conference on
Conference_Location :
Antalya
Print_ISBN :
978-1-4577-0462-8
Electronic_ISBN :
978-1-4577-0461-1
DOI :
10.1109/SIU.2011.5929712
