Speaker Recognition Using a Binary Representation and Specificities Models

Hernández-Sierra, Gabriel; Bonastre, Jean-François; Lara, José Ramón Calvo de

doi:10.1007/978-3-642-33275-3_90

Cited by 6 publications

(22 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to apply standardized biometric template protection schemes, binarization can be employed [3]. Related work on the binarization of traditional speaker recognition systems utilizing universal background models (UBMs) targeting the GMM -UBM approach can be found in [25,26,27]. In addition, in our earlier work [28], we proposed a biometric template protection scheme for speaker recognition, based on binarized Gaussian mixture model (GMM) supervectors.…”

Section: Related Workmentioning

confidence: 99%

Homomorphic Encryption for Speaker Recognition: Protection of Biometric Templates and Vendor Model Parameters

Nautsch¹,

Isadskiy²,

Kolberg³

et al. 2018

The Speaker and Language Recognition Workshop (Odyssey 2018)

View full text Add to dashboard Cite

Data privacy is crucial when dealing with biometric data. Accounting for the latest European data privacy regulation and payment service directive, biometric template protection is essential for any commercial application. Ensuring unlinkability across biometric service operators, irreversibility of leaked encrypted templates, and renewability of e.g., voice models following the i-vector paradigm, biometric voice-based systems are prepared for the latest EU data privacy legislation. Employing Paillier cryptosystems, Euclidean and cosine comparators are known to ensure data privacy demands, without loss of discrimination nor calibration performance. Bridging gaps from template protection to speaker recognition, two architectures are proposed for the two-covariance comparator, serving as a generative model in this study. The first architecture preserves privacy of biometric data capture subjects. In the second architecture, model parameters of the comparator are encrypted as well, such that biometric service providers can supply the same comparison modules employing different key pairs to multiple biometric service operators. An experimental proof-of-concept and complexity analysis is carried out on the data from the 2013 -2014 NIST i-vector machine learning challenge.

show abstract

Section: Related Workmentioning

confidence: 99%

Homomorphic Encryption for Speaker Recognition: Protection of Biometric Templates and Vendor Model Parameters

Nautsch¹,

Isadskiy²,

Kolberg³

et al. 2018

The Speaker and Language Recognition Workshop (Odyssey 2018)

View full text Add to dashboard Cite

show abstract

“…Those specificities are obtained from the adapted models of the training set, by matching the components of the adapted models to the component of the UBM. Since the specificities number is assumed to be very large, it is necessary to reduce it, selecting the most important [4]. As a result the number of specificities per acoustic class could not be the same.…”

Section: Accumulative Vectorsmentioning

confidence: 99%

“…Then the components with the highest probability were selected, the specificities of these components are the ones represented in the binary vector. We use 3 based on previous results presented in [4]. Then for each component is compute the likelihood of each acoustic frame with all the corresponding specificities.…”

Section: Accumulative Vectorsmentioning

confidence: 99%

“…It deals directly with the speaker discriminant information in a discrete and binary space. Our method to obtain the binary representation and then the accumulative vectors is similar to [4], it only differs on the normalization process used for reducing the susceptibility of the accumulative vectors to the signal length. At this point we introduce a new method for successfully accomplish this task since it shows better performance combined with the SVM.…”

Section: Introductionmentioning

confidence: 99%

“…This paper is organized as follows. Section 2 explains the process to obtain the accumulative vectors presented in [4]. Section 3 briefly describes SVM paradigm and NAP algorithm.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Speaker Verification Using Accumulative Vectors with Support Vector Machines

Martínez

Hernández-Sierra

Lara

2013

Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications

Self Cite

View full text Add to dashboard Cite

The applications of Support Vector Machines (SVM) in speaker recognition are mainly related to Gaussian Mixtures and Universal Background Model based supervector paradigm. Recently, has been proposed a new approach that allows represent each acoustic frame in a binary discriminant space. Also a representation of a speaker-called accumulative vectors-obtained from the binary space has been proposed. In this article we show results obtained using SVM with the accumulative vectors and Nuisance Attribute Projection (NAP) as a method for compensating the session variability. We also introduce a new method to counteract the effects of the signal length in the conformation of the accumulative vectors to improve the performance of SVM.

show abstract

Temporal Information in a Binary Framework for Speaker Recognition

Hernández-Sierra

Calvo

Bonastre

2014

Advanced Information Systems Engineering

Self Cite

View full text Add to dashboard Cite

In recent years a simple representation of a speech excerpt has been proposed, as a binary matrix allowing easy access to the speaker discriminant information. In addition to the time-related abilities of this representation, it also allows the system to work with a temporal information representation based on sequential changes present in the binary representation. A new temporal information is proposed in order to add it to speaker recognition systems. A new specificity selection approach using a mask in the cumulative vector space is also proposed. This aims to increase effectiveness in the speaker binary key paradigm. The experimental validation, done on the NIST-SRE framework, demonstrates the efficiency of the proposed solutions, which shows an EER improvement of 7%. The combination of i-vector and binary approaches, using the proposed methods, showed the complementarity of the discriminatory information exploited by each of them.

show abstract

Speaker Recognition Using a Binary Representation and Specificities Models

Cited by 6 publications

References 6 publications

Homomorphic Encryption for Speaker Recognition: Protection of Biometric Templates and Vendor Model Parameters

Homomorphic Encryption for Speaker Recognition: Protection of Biometric Templates and Vendor Model Parameters

Speaker Verification Using Accumulative Vectors with Support Vector Machines

Temporal Information in a Binary Framework for Speaker Recognition

Contact Info

Product

Resources

About