Exploiting sequence information for text-dependent Speaker Verification

Dey, Subhadeep; Motlíček, Petr; Madikeri, Srikanth; Ferras, Marc

doi:10.1109/icassp.2017.7953182

Cited by 9 publications

(8 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our previous work [12,16], we used online i-vectors as features to Dynamic Time Warping (DTW) algorithm for fixed phrase based text-dependent SV task. Significant gain in performance was observed as opposed to using the conventional i-vectors which suggests that these features contain sufficient speaker and content information.…”

Section: Content Normalization Using I-vectorsmentioning

confidence: 99%

Content Normalization for Text-Dependent Speaker Verification

et al. 2017

Self Cite

View full text Add to dashboard Cite

Subspace based techniques, such as i-vector and Joint Factor Analysis (JFA) have shown to provide state-of-the-art performance for fixed phrase based text-dependent speaker verification. However, the error rates of such systems on the random digit task of RSR dataset are higher than that of Gaussian Mixture Model-Universal Background Model (GMM-UBM). In this paper, we aim at improving i-vector system by normalizing the content of the enrollment data to match the test data. We estimate i-vectors for each frames of a speech utterance (also called online i-vectors). The largest similarity scores across frames between enrollment and test are taken using these online i-vectors to obtain speaker verification scores. Experiments on Part3 of RSR corpora show that the proposed approach achieves 12% relative improvement in equal error rate over a GMM-UBM based baseline system.

show abstract

Section: Content Normalization Using I-vectorsmentioning

confidence: 99%

Content Normalization for Text-Dependent Speaker Verification

et al. 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, a limitation of the d-Vector and triplet-loss based approaches is that they ignore the content-information of the speech signal completely. Many studies have shown that exploiting the phonetic variability can significantly enhance the performance of SV systems [5,4,13,14,15]. Motivated by these results, we incorporated this information on top of DNN based speaker embedding (from the d-Vector and tripletloss network) for the Random-digit task using content-matching [5].…”

Section: Introductionmentioning

confidence: 99%

DNN Based Speaker Embedding Using Content Information for Text-Dependent Speaker Verification

Dey

Koshinaka²,

Motlíček

et al. 2018

2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

In this paper, we are interested in exploring Deep Neural Network (DNN) based speaker embedding for Random-digit task using content information. To this end, a technique is applied to automatically select common phonetic units between the enrollment and test data to produce speaker verification scores. Furthermore, a novel approach is proposed to incorporate content information in the DNN directly. It is hypothesized that features extracted using this DNN will be helpful for the task. Experiments on the RSR dataset show that the proposed method outperforms the baseline i-vector system by 43% relative equal error rate.

show abstract

“…Text-dependent SV can be implemented in various ways [7,6,8,9,10] (such as phrase, seen-content, random-digit, or short commands -based authentication). In this paper, we are interested in fixed-phrase and random-digit type of text-dependent SV systems.…”

Section: Introductionmentioning

confidence: 99%

End-to-end Text-dependent Speaker Verification Using Novel Distance Measures

2018

Self Cite

View full text Add to dashboard Cite

This paper explores novel ideas in building end-to-end deep neural network (DNN) based text-dependent speaker verification (SV) system. The baseline approach consists of mapping a variable length speech segment to a fixed dimensional speaker vector by estimating the mean of hidden representations in DNN structure. The distance between two utterances is obtained by computing L2 norm between the vectors. This approach performs worse than the conventional Gaussian Mixture Model-Universal Background Model (GMM-UBM) based SV on a publicly available corpora. We believe that a degraded performance is due to the employed averaging operation, which may not capture the phonetic information of an utterance. Recent studies indicate that techniques exploiting phonetic information in addition to speaker is beneficial for this task. This paper therefore proposes to incorporate content information of the speech signal by computing distance function with linguistic units co-occuring between enrollment and test data. The whole network is optimized by employing a triplet-loss objective in an end-to-end fashion to estimate SV scores. Experiments on the RSR2015 dataset indicate that the proposed approach outperforms GMM-UBM system by 48% and 36% relative equal error rate for fixed-phrase and random-digit conditions respectively.

show abstract

Exploiting sequence information for text-dependent Speaker Verification

Cited by 9 publications

References 19 publications

Content Normalization for Text-Dependent Speaker Verification

Content Normalization for Text-Dependent Speaker Verification

DNN Based Speaker Embedding Using Content Information for Text-Dependent Speaker Verification

End-to-end Text-dependent Speaker Verification Using Novel Distance Measures

Contact Info

Product

Resources

About