ISV-based speaker verification in spoof scenario on presentation attacks of AVspoof dataset

Outputs for block analysis

eer 0.0479927
far_dev 0.0479381
far_test 0.97747
frr_dev 0.0480473
frr_test 0.0314359
hter 0.504453
threshold 0.485038
number_of_negatives_dev 54925
number_of_negatives_test 43320
number_of_positives_dev 4225
number_of_positives_test 4708
roc_dev
roc_test
To edit a toolchain, please use a modern browser (Mozilla Firefox 3.0+, Google Chrome 1+, Apple Safari 3+, Opera 9.5+, Microsoft Internet Explorer 9+)

This is an automatic speaker verification (ASV) system based on ISV modeling for a spoof scenario. Only presentation attacks from AVspoof database are used. GMM with 256 Gaussians was used to train a UBM. This experiment is essentially the same system as in pkorshunov/isv-speaker-verification/1/isv-speaker-verification-licit, but when spoofing data is used to attack the system.

Experiments which are referred by reports or by an associated attestation cannot be modified. You may revert the situation by deleting unpublished reports or the attestation in case they have not been published. Note you may still fork this experiment and re-run a modified or unmodified version of it.

This experiment is not yet attested.
Terms of Service | Contact Information | BEAT platform version 1.3.1rc1 | © Idiap Research Institute - 2013-2019