Loading...
report
Infinite Models for Speaker Clustering
Valente, Fabio
2006
In this paper we propose the use of infinite models for the clustering of speakers. Speaker segmentation is obtained trough a Dirichlet Process Mixture (DPM) model which can be interpreted as a flexible model with an infinite a priori number of components. Learning is based on a Variational Bayesian approximation of the infinite sequence. DPM model is compared with fixed prior systems learned by ML/BIC, MAP/BIC and a Variational Bayesian method. Experiments are run on a speaker clustering task on the NIST-96 Broadcast News database.
Loading...
Name
valente-idiap-rr-06-19.pdf
Access type
openaccess
Size
122.44 KB
Format
Adobe PDF
Checksum (MD5)
b14bfc696046ea87005f404f353b68bd