i-vector/HMM Based Text-dependent Speaker Verification System for RedDots Challenge

• https://www.researchgate.net/publication/303895014_i-VectorHMM_Based_Text-Dependent_Speaker_Verification_System_for_RedDots_Challenge
• http://www.fit.vutbr.cz/research/groups/speech/publi/2016/zeinali_interspeech2016_IS161174.pdf PDF

text-dependent speaker verification, i-vector, HMM, RedDots challenge

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.

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.