A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2018; you can also visit the original URL.
The file type is application/pdf
.
Filters
Factor Analysis of Neighborhood-Preserving Embedding for Speaker Verification
2012
IEICE transactions on information and systems
In this letter, we adopt a new factor analysis of neighborhood-preserving embedding (NPE) for speaker verification. ...
NPE aims at preserving the local neighborhood structure on the data and defines a low-dimensional speaker space called neighborhood-preserving embedding space. ...
Conclusion In this letter, we propose a new factor analysis method of neighborhood-preserving embedding (NPE) to speaker verification. ...
doi:10.1587/transinf.e95.d.2572
fatcat:qkp6oxko6ne2lf5mn3p3r2foxe
Locality Preserving Discriminant Projection for Speaker Verification
2020
Journal of Computer and Communications
In this paper, a manifold subspace learning algorithm based on locality preserving discriminant projection (LPDP) is used for speaker verification. ...
LPDP can overcome the deficiency of the total variability factor analysis and locality preserving projection (LPP). LPDP can effectively use the speaker label information of speech data. ...
Conflicts of Interest The authors declare no conflicts of interest regarding the publication of this paper. ...
doi:10.4236/jcc.2020.811002
fatcat:ckhfyfhalfa2hfexzhpm4cxglm
Speaker Verification Using Secure Binary Embeddings
2013
Zenodo
Publication in the conference proceedings of EUSIPCO, Marrakech, Morocco, 2013 ...
In this paper we propose a new privacy-preserving technique for speaker verification based on the recently proposed Secure Binary Embeddings (SBE) [4] . ...
These variants typically employ various flavors of factor analysis [6] [7] to assign a priori probabilities to the parameters of the GMM for the speaker, and highly accurate authentication is reported ...
doi:10.5281/zenodo.43432
fatcat:t5s6setkj5aidgyrqhyu6qfdiu
Using the conformal embedding analysis to compensate the channel effect in the i-vector based speaker verification system
2013
Biometrics and Electronic Signatures
computational complexity.I nt his paper,w ep ropose to use Conformal Embedding Analysis (CEA ); arecently proposed manifold leaning technique; to tackle the main limitations of LDAw hich are: the Gaussian ...
Experimental results on the challenging MOBIO-voice database showt hat CEA+WCCN outperforms LDA+WCCN for both male and female speakers at all operating points. ...
and Neighborhood Preserving Embedding (NPE) [HCYZ05] . ...
dblp:conf/biosig/BoulkenafetBNC13
fatcat:x6dkzr7n6vempk2xpq3bvaztmy
Improved Supervised Locality Preserving Projection for I-vector Based Speaker Verification
2018
Interspeech 2018
A Supervised Locality Preserving Projection (SLPP) method is employed for channel compensation in an i-vector based speaker verification system. ...
Index Terms: speaker verification, supervised locality preserving projection, probabilistic linear discriminant analysis ...
Results and analysis
Conclusions In this study, we employ SLPP for channel compensation in an i-vector based speaker verification system. ...
doi:10.21437/interspeech.2018-41
dblp:conf/interspeech/YouGSZ18
fatcat:q3eeyhhlijcophhyll4gp27fy4
Deep Latent Space Learning for Cross-modal Mapping of Audio and Visual Signals
[article]
2019
arXiv
pre-print
Our experiments demonstrate the effectiveness of the technique for cross-modal biometric applications. ...
State-of-the-art performance is achieved on cross-modal verification and matching while comparable results are observed on the remaining applications. ...
It thus imposes neighborhood preserving constraint within each modality as well as across modalities. ...
arXiv:1909.08685v1
fatcat:f5zydb2dmreqzdzcpndsqneq7u
Graph Embedding for Speaker Recognition
[chapter]
2012
Graph Embedding for Pattern Analysis
In addition, methods such as iVector and joint factor analysis can be written in this form, since factor analysis can be written as a linear transform-see [16] for more details. ...
[30] , to obtain the low-dimensional embedding that best preserves these distances. ...
doi:10.1007/978-1-4614-4457-2_10
fatcat:vbocs3buijbkplcrgdaozavgqq
Privacy-Preserving Adversarial Representation Learning in ASR: Reality or Illusion?
2019
Interspeech 2019
Through speaker identification and verification experiments on the Librispeech corpus with open and closed sets of speakers, we show that the representations obtained from a standard architecture still ...
protection of the speaker identity in practice. ...
This increased the amount of data by a factor of 4. ...
doi:10.21437/interspeech.2019-2415
dblp:conf/interspeech/SrivastavaBTV19
fatcat:poxxcsqvcfcv5flrzwaz6fxxzm
Privacy-Preserving Speaker Authentication
[chapter]
2012
Lecture Notes in Computer Science
Speaker authentication systems require access to the voice of the user. ...
In this paper we present a new technique that employs secure binary embeddings of feature vectors, to perform voice authentication in a privacy preserving manner with minimal computational overhead and ...
The authors would like to thank Petros Boufounos for many helpful suggestions. ...
doi:10.1007/978-3-642-33383-5_1
fatcat:7blmpaf4xfcrjmxkynavzgmyta
Audio-Visual Kinship Verification
[article]
2019
arXiv
pre-print
In this paper, we address, for the first time, the use of audio-visual information from face and voice modalities to perform kinship verification. ...
Then, early and late fusion methods are evaluated on the TALKIN dataset for the study of kinship verification with both face and voice modalities. ...
Hadid for their technical advice for this paper. ...
arXiv:1906.10096v1
fatcat:stoetshcynb5zlryni34i2dosm
Audio-Visual Kinship Verification in the Wild
2019
2019 International Conference on Biometrics (ICB)
State-of-the-art conventional and deep learning models are assessed and compared for kinship verification using this dataset. ...
Finally, we propose a deep Siamese network for multi-modal fusion of kinship relations. ...
Acknowledgment This work is partially supported by the China Scholarship Council (grant 201706290103), the Academy of Finland, and the Natural Sciences and Engineering Research Council of Canada. ...
doi:10.1109/icb45273.2019.8987241
dblp:conf/icb/WuGKFH19
fatcat:dwaccvbsfjcollypzjmmzxuyx4
Centroid-aware local discriminative metric learning in speaker verification
2017
Pattern Recognition
Abstract We propose a new mechanism to pave the way for efficient learning against class-imbalance and improve representation of identity vector (i-vector) in automatic speaker verification (ASV). ...
Centroid-aware local discriminative metric learning in speaker verification.. Pattern Recognition, Elsevier, 2017, 72 (c), pp. ...
for meaningful discussions and help. Besides, the authors thank all reviewers for their valuable comments on improving the quality of this paper. ...
doi:10.1016/j.patcog.2017.07.007
fatcat:7u5kynatpfgexg7flygk5e27ei
Graph-Based Semisupervised Learning for Acoustic Modeling in Automatic Speech Recognition
2016
IEEE/ACM Transactions on Audio Speech and Language Processing
More importantly, the performance of the acoustic model could degenerate during test time, where the conditions of test data differ from the training data in speaker characteristics, channel and recording ...
However, lots of labor is required to annotate the training data for automatic speech recognition. ...
Figure 6 . 6 : 66 Procedure for extracting graph-embedding features.
Figure 6 . 7 : 67 Procedure for extracting speaker-dependent neighborhood graph embedding features. ...
doi:10.1109/taslp.2016.2593800
fatcat:xysvty354vc2ddk2kk5xrhs6gu
Soft computing and intelligent systems: Techniques and applications
2018
Journal of Intelligent & Fuzzy Systems
For preserving data confidentiality, the client employs few linear transformation schemes. The proposed verification scheme helps to maintain the integrity of the computed result. ...
This paper also provides a comparative analysis with other machine-learning classifiers. ...
For preserving data confidentiality, the client employs few linear transformation schemes. The proposed verification scheme helps to maintain the integrity of the computed result. ...
doi:10.3233/jifs-169420
fatcat:cm65pxnfxffatckhhgsbfw2vje
Self-supervised Representation Learning With Path Integral Clustering For Speaker Diarization
[article]
2021
arXiv
pre-print
In addition, the DER results reported in this work improve over several other recent approaches for speaker diarization. ...
In this paper, we propose a representation learning and clustering algorithm that can be iteratively performed for improved speaker diarization. ...
The model based affinity scoring using probabilistic linear discriminant analysis (PLDA) [32] , which was successful in speaker verification [33] , has also been incorporated for speaker diarization ...
arXiv:2104.09456v1
fatcat:qgdfw3eiabdijlkhftja2h4tte
« Previous
Showing results 1 — 15 out of 1,568 results