Publication Details
i-vector/HMM Based Text-dependent Speaker Verification System for RedDots Challenge
Sameti Hossein
Burget Lukáš, doc. Ing., Ph.D. (DCGM)
Černocký Jan, prof. Dr. Ing. (DCGM)
Maghsoodi Nooshin
Matějka Pavel, Ing., Ph.D.
text-dependent speaker verification, i-vector,HMM, RedDots challenge
Recently, a new data collection was initiated within the RedDotsproject in order to evaluate text-dependent and text-promptedspeaker recognition technology on data from a wider speakerpopulation and with more realistic noise, channel and phoneticvariability. This paper analyses our systems built for RedDotschallenge - the effort to collect and compare the initial resultson this new evaluation data set obtained at different sites.We use our recently introduced HMM based i-vector approach,where, instead of the traditional GMM, a set of phone specificHMMs is used to collect the sufficient statistics for i-vectorextraction. Our systems are trained in a completely phraseindependentway on the data from RSR2015 and Libri speechdatabases. We compare systems making use of standard cepstralfeatures and their combination with neural network basedbottle-neck features. The best results are obtained with a scorelevelfusion of such systems.
Recently, a new data collection was initiated within the RedDots project in order to evaluate text-dependent and text-prompted speaker recognition technology on data from a wider speaker population and with more realistic noise, channel and phonetic variability. This paper analyses our systems built for RedDots challenge - the effort to collect and compare the initial results on this new evaluation data set obtained at different sites. We use our recently introduced HMM based i-vector approach, where, instead of the traditional GMM, a set of phone specific HMMs is used to collect the sufficient statistics for i-vector extraction. Our systems are trained in a completely phraseindependent way on the data from RSR2015 and Libri speech databases. We compare systems making use of standard cepstral features and their combination with neural network based bottle-neck features. The best results are obtained with a scorelevel fusion of such systems.
@inproceedings{BUT131018,
author="Hossein {Zeinali} and Hossein {Sameti} and Lukáš {Burget} and Jan {Černocký} and Nooshin {Maghsoodi} and Pavel {Matějka}",
title="i-vector/HMM Based Text-dependent Speaker Verification System for RedDots Challenge",
booktitle="Proceedings of Interspeech 2016",
year="2016",
pages="440--444",
publisher="International Speech Communication Association",
address="San Francisco",
doi="10.21437/Interspeech.2016-1174",
isbn="978-1-5108-3313-5",
url="https://www.researchgate.net/publication/303895014_i-VectorHMM_Based_Text-Dependent_Speaker_Verification_System_for_RedDots_Challenge"
}