Please use this identifier to cite or link to this item:
Title: Behavior of a bayesian adaptation method for incremental enrollment in speaker verification
Authors: Fredouille , Corinne
Mariethoz, J
Jaboulet , C
Mokbel, Chafic 
Affiliations: Department of Electrical Engineering 
Keywords: Bayesian methods
Hidden Markov models
Context modeling
Speech recognition
Speaker recognition
Covariance matrix
Viterbi algorithm
Issue Date: 2000
Part of: 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing.
Start page: 1
End page: 4
Conference: International Conference on Acoustics, Speech, and Signal Processing (5-9 June 2000 : Istanbul, Turkey,) 
Classical adaptation approaches are generally used for speaker or environment adaptation of speech recognition systems. In this paper, we use such techniques for the incremental training of client models in a speaker verification system. The initial model is trained on a very limited amount of data and then progressively updated with access data, using a segmental-EM procedure. In supervised mode (i.e. when access utterances are certified), the incremental approach yields equivalent performance to the batch one. We also investigate on the impact of various scenarios of impostor attacks during the incremental enrollment phase. All results are obtained with the Picassoft platform-the state-of-the-art speaker verification system developed in the PICASSO project.
Ezproxy URL: Link to full text
Type: Conference Paper
Appears in Collections:Department of Electrical Engineering

Show full item record

Record view(s)

checked on Apr 22, 2024

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.