DNN-Driven Driven Mixture of PLDA for Robust Speaker Verification
Abstract: The mismatch between enrollment and test utterances due to different types of variabilities is a great challenge in speaker verification. Based on the observation that the SNR-level level variability or channel channel-type type variability causes heterogeneous clusters in i-vector vector space, this paper proposes to apply supervised learning to drive or guide the learning ing of probabilistic linear discriminant analysis (PLDA) mixture models. Specifically, a deep neural network (DNN) is trained to produce the posterior probabilities of different SNR levels or channel types given i-vectors i as input. These posteriors then re replace place the posterior probabilities of indicator variables in the mixture of PLDA. The discriminative training causes the mixture model to perform more reasonable soft divisions of the ii-vector vector space as compared to the conventional mixture of PLDA. During ve verification, rification, given a test ii vector and a target-speaker's speaker's ii-vector, vector, the marginal likelihood for the samesame speaker hypothesis is obtained by summing the component likelihoods weighted by the component posteriors produced by the DNN, and likewise for the different-speaker speaker hypothesis. Results based on NIST 2012 SRE demonstrate that the proposed scheme leads to better performance under more realistic situations where both training and test utterances cover a wide range of SNRs and different channel types. Unlike the previous SNR SNR-dependent dependent mixture of PLDA which only focuses on SNR mismatch, the proposed model is more general and is potentially applicable to addressing different types of variability in speech.