Muutke küpsiste eelistusi

E-raamat: Voice Biometrics: Technology, trust and security

Edited by (University of Vigo, Department of Signal Theory and Communications, Spain), Edited by (Intelligent Voice, UK)
  • Formaat: EPUB+DRM
  • Sari: Security
  • Ilmumisaeg: 29-Oct-2021
  • Kirjastus: Institution of Engineering and Technology
  • Keel: eng
  • ISBN-13: 9781785619014
Teised raamatud teemal:
  • Formaat - EPUB+DRM
  • Hind: 195,00 €*
  • * hind on lõplik, st. muud allahindlused enam ei rakendu
  • Lisa ostukorvi
  • Lisa soovinimekirja
  • See e-raamat on mõeldud ainult isiklikuks kasutamiseks. E-raamatuid ei saa tagastada.
Voice Biometrics: Technology, trust and securitySuurem pilt
  • Formaat: EPUB+DRM
  • Sari: Security
  • Ilmumisaeg: 29-Oct-2021
  • Kirjastus: Institution of Engineering and Technology
  • Keel: eng
  • ISBN-13: 9781785619014
Teised raamatud teemal:

DRM piirangud

  • Kopeerimine (copy/paste):

    ei ole lubatud

  • Printimine:

    ei ole lubatud

  • Kasutamine:

    Digitaalõiguste kaitse (DRM)
    Kirjastus on väljastanud selle e-raamatu krüpteeritud kujul, mis tähendab, et selle lugemiseks peate installeerima spetsiaalse tarkvara. Samuti peate looma endale  Adobe ID Rohkem infot siin. E-raamatut saab lugeda 1 kasutaja ning alla laadida kuni 6'de seadmesse (kõik autoriseeritud sama Adobe ID-ga).

    Vajalik tarkvara
    Mobiilsetes seadmetes (telefon või tahvelarvuti) lugemiseks peate installeerima selle tasuta rakenduse: PocketBook Reader (iOS / Android)

    PC või Mac seadmes lugemiseks peate installima Adobe Digital Editionsi (Seeon tasuta rakendus spetsiaalselt e-raamatute lugemiseks. Seda ei tohi segamini ajada Adober Reader'iga, mis tõenäoliselt on juba teie arvutisse installeeritud )

    Seda e-raamatut ei saa lugeda Amazon Kindle's. 

Voice biometrics are being implemented globally in large scale applications such as remote banking, government e-services, transportation and building security access, autonomous vehicles, and healthcare. They have been integrated in numerous apps, often coupled with face biometrics and artificial intelligence methods. Voice biometrics products and solutions must meet three key requirements for the success in their deployment: they must be highly trustable regarding privacy protection; easy to use and always be available.



This edited book presents the state of the art in voice biometrics research and technologies including implementation and deployment challenges in terms of interoperability, scalability and performance, and security. The team of editors and chapter authors combine a wealth of expertise from academia and the industry. Topics covered include the fundamentals of voice biometrics; design of countermeasures for replay attack; attacker's perspective for voice biometrics; voice biometrics; speaker de-identification; performance evaluation of voice biometrics solutions; standardization of voice biometrics technology; industry perspectives; joining forces of voice and facial biometrics; and future trends and challenges in voice biometrics.



Providing comprehensive coverage of the field of voice biometrics, this authoritative volume will be of great interest to researchers, scientists, engineers, practitioners and advanced students involved in the fields of security, biometrics, forensic sciences, human computer interaction, speech processing, acoustics, multimedia, pattern recognition, and privacy-preserving, digital signal processing and speech technologies. It will also be of interest to researchers and professionals working in law and criminology.
List of figures
xi
List of tables
xv
Short biographies of the editors and authors xvii
Preface to Voice Biometrics xxvii
About the editors xxix
1 Introduction
1(1)
Carmen Garcia-Mateo
Gerard Chollet
Chapter 2 Fundamentals of voice biometrics: classical and machine learning approaches
1(1)
Chapter 3 Voice biometrics: attacker's perspective
2(1)
Chapter 4 Voice biometrics: privacy in paralinguistic and extralinguistic tasks for health applications
3(1)
Chapter 5 Voice privacy in biometrics: speaker de-identification
3(1)
Chapter 6 Performance evaluation of voice biometrics solutions
3(1)
Chapter 7 Voice biometrics: how the technology is standardized
4(1)
Chapter 8 Voice biometrics: perspective from the industry
4(1)
Chapter 9 Joining forces of voice and facial biometrics: a case study in the scope of NIST SRE' 19
4(1)
Chapter 10 Voice biometrics: future trends and challenges ahead
5(2)
2 Fundamentals of voice biometrics: classical and machine learning approaches
7(32)
Alicia Lozano-Diez
Joaquin Gonzalez-Rodriguez
Daniel Ramos
Doroteo T. Toledano
2.1 Introduction to speaker recognition systems
7(2)
2.2 Metrics for system performance evaluation
9(2)
2.2.1 ROC, DET and EER
9(1)
2.2.2 Detection cost function
10(1)
2.3 Text-independent speaker recognition
11(10)
2.3.1 Classical acoustic approaches: GMM-UBM, i-vector and PLDA
11(3)
2.3.2 DNN approaches
14(1)
2.3.2.1 Basic concepts of neural networks
14(3)
2.3.2.2 Some applications of DNNs to speech processing
17(1)
2.3.3 DNNs for speaker recognition
17(4)
2.4 Text-dependent speaker recognition
21(2)
2.4.1 Classification of systems and techniques
21(1)
2.4.2 Databases and benchmarks
22(1)
2.5 Calibration of speaker recognition scores
23(16)
2.5.1 Motivation: why to calibrate?
23(2)
2.5.2 What is calibration?
25(2)
2.5.3 Score-to-LR computation methods
27(1)
2.5.3.1 Generative calibration models: fitting distributions to scores
28(1)
2.5.3.2 Discriminative calibration models: transforming scores into LR values to optimize a cost function
28(2)
2.5.4 Performance measurement of score-to-LR methods
30(1)
References
31(8)
3 Voice biometrics: attacker's perspective
39(28)
Priyanka Gupta
Hemant A. Patil
Abstract
39(1)
3.1 Introduction
40(2)
3.2 Direct attacks
42(6)
3.2.1 Spoofing attacks
42(2)
3.2.2 Black box hardware attacks
44(1)
3.2.3 Black box adversarial attacks
45(3)
3.3 Indirect attacks
48(5)
3.3.1 Attacks on corpora
48(3)
3.3.2 Gray box hardware attacks
51(1)
3.3.3 Gray box and white box adversarial attacks
51(2)
3.4 Technological challenges
53(4)
3.4.1 Extracting prosodic information
53(1)
3.4.2 Enrolled users with malicious intent
53(1)
3.4.3 Number of trials permitted on the ASV
54(1)
3.4.4 Minuteness of the perturbation in adversarial attacks
54(1)
3.4.5 Privacy preservation of speech and voice privacy
55(2)
3.5 Conclusions and future work
57(10)
Acknowledgments
57(1)
References
58(9)
4 Voice biometrics: privacy in paralinguistic and extralinguistic tasks for health applications
67(26)
Francisco Teixeira
Alberto Abad
Isabel Trancoso
Bhiksha Raj
4.1 Introduction
67(2)
4.2 Paralinguistic and extralinguistic tasks
69(3)
4.2.1 Speech-affecting diseases
70(1)
4.2.2 Methods
71(1)
4.3 Cryptographic primitives and MPC for PPML
72(6)
4.3.1 Homomorphic encryption
72(1)
4.3.2 Oblivious transfer
73(1)
4.3.3 Secure Multiparty Computation
73(1)
4.3.3.1 Yao's GCs protocol
74(1)
4.3.3.2 Secret sharing
75(1)
4.3.3.3 Security models
76(1)
4.3.4 Distance-preserving hashing techniques
77(1)
Secure binary embeddings
77(1)
Secure modular hashing
78(1)
4.4 PPML for paralinguistic and extralinguistic tasks
78(8)
4.4.1 PPML for non-health-related tasks
78(1)
4.4.2 PPML for health-related tasks
79(1)
4.4.3 Private SVM+RBF for health-related tasks
80(1)
4.4.3.1 Private RBF computation
81(1)
4.4.3.2 Private SVM computation
81(1)
4.4.3.3 Experimental setup
82(1)
4.4.3.4 Model training and parameters
83(1)
4.4.3.5 Private SVM implementation details
83(1)
4.4.3.6 Classification results
84(1)
4.4.3.7 Security and computational performance
84(2)
4.5 Conclusions
86(7)
Acknowledgements
86(1)
References
86(7)
5 Voice privacy in biometrics: speaker de-identification
93(28)
Paula Lopez-Otero
Laura Docio-Fernandez
Carmen Garcia-Mateo
5.1 Introduction
93(2)
5.2 How to evaluate speaker de-identification?
95(3)
5.2.1 Subjective measures
95(2)
5.2.2 Objective measures
97(1)
5.3 Speaker de-identification techniques
98(4)
5.3.1 Codebook mapping
99(1)
5.3.2 Gaussian mixture model
100(1)
5.3.3 Frequency warping
101(1)
5.3.4 Deep learning techniques
102(1)
5.4 Experiment definition
102(6)
5.4.1 Piecewise definition of transformation functions
103(1)
5.4.2 Pretrained transformation functions
104(2)
5.4.3 De-identification based on DNNs
106(1)
5.4.4 De-identification based on generative adversarial networks
107(1)
5.5 Evaluation corpora
108(3)
5.5.1 Evaluation metrics
109(2)
5.6 Results and analysis
111(2)
5.7 Conclusion
113(8)
Acknowledgements
114(1)
References
114(7)
6 Performance evaluation of voice biometrics solutions
121(18)
Jean-Francois Bonastre
Anthony Larcher
6.1 Introduction
121(2)
6.2 Evaluating methods or technology
123(8)
6.2.1 Existing benchmarking evaluations
124(1)
6.2.2 Evaluation criteria
125(1)
6.2.2.1 Evaluating a system producing hard decisions
125(2)
6.2.2.2 Evaluating the goodness of verification scores
127(2)
6.2.3 Statistical significance
129(1)
6.2.4 Specific evaluation aspects
129(1)
6.2.5 Evaluating related technologies
130(1)
6.3 Bias in testing
131(1)
6.4 Summary and propositions
132(7)
References
133(6)
7 Voice biometrics: How the technology is standardized
139(24)
Andreas Nautsch
Christoph Busch
7.1 Introduction
140(1)
7.2 Biometrics standardization within ISO/IEC
140(9)
7.2.1 Generalized system design
141(2)
7.2.2 Harmonized biometric vocabulary
143(1)
7.2.3 Performance testing and reporting
144(2)
7.2.4 Presentation attack detection
146(1)
7.2.5 Biometric information protection
147(2)
7.3 Data interchange formats for passports and beyond
149(5)
7.3.1 Motivation and background on encoding biometric data
150(1)
7.3.2 Data interchange standard ISO/IEC 19794
151(1)
7.3.3 Format structure
152(2)
7.3.4 ISO/IEC 19794 Part 13: voice data
154(1)
7.4 Discussion: de facto and ISO/IEC standards
154(5)
7.4.1 On the general system design
154(2)
7.4.2 Gap analysis: performance testing and reporting
156(2)
7.4.3 Regarding implementations and data interchange formats
158(1)
7.5 Conclusion
159(4)
Acknowledgements
160(1)
References
160(3)
8 Voice biometrics: perspective from the industry
163(23)
Marcel Kockmann
Kevin Farrell
Daniele Colibro
Claudio Vair
Anil Alexander
Finnian Kelly
8.1 Automated password reset: an example of a commercial application using voice biometrics
164(5)
8.1.1 Overview
164(1)
8.1.2 Introduction
164(1)
8.1.3 System architecture
165(2)
8.1.4 Voice biometric system
167(2)
8.1.5 Summary
169(1)
8.2 Testing of commercial voice biometric systems
169(6)
8.2.1 Introduction
169(1)
8.2.1.1 Biometric testing
170(2)
8.2.2 User analysis
172(2)
8.2.3 Summary
174(1)
8.3 Forensic speaker recognition
175(11)
8.3.1 Introduction
175(1)
8.3.2 Forensic speaker recognition and the strength of evidence
176(1)
8.3.3 The forensic expert's workflow
176(2)
8.3.4 Technical challenges
178(2)
8.3.4.1 Improving interpretability of scores
180(1)
8.3.4.2 Score normalization
180(1)
8.3.4.3 Score calibration
181(1)
8.3.4.4 Condition adaptation
181(1)
8.3.4.5 Dealing with multi-speaker recordings
182(1)
8.3.5 Training-communication between system developers and end-users
182(1)
8.3.6 Conclusions
182(1)
References
183(3)
9 Joining forces of voice and facial biometrics: a case study in the scope of NI ST SRE'19
186(33)
Mohamed Amine Hmani
Aymen Mtibaa
Dijana Petrovska Delacretaz
9.1 Introduction to the NIST SRE' 19 challenge
188(2)
9.1.1 The SRE' 19 CTS challenge
188(1)
9.1.2 The SRE' 19 multimedia challenge
188(1)
9.1.3 SRE'19 evaluation metrics
189(1)
9.2 TSP speaker verification system for the SRE' 19 evaluation
190(7)
9.2.1 A brief review of state of the art in speaker verification
190(1)
9.2.2 TSP speaker verification common pipeline for the SRE'
190(1)
CTS and multimedia challenges
191(1)
TDNN
191(1)
E-TDNN
192(1)
9.2.3 TSP speaker verification system for the SRE' 19 CTS challenge
192(1)
9.2.4 TSP speaker verification system for the SRE' 19 multimedia challenge
193(1)
9.2.5 Results for TSP speaker verification systems on the SRE' 19 CTS and multimedia challenges
194(3)
9.2.6 Conclusions
197(1)
9.3 TSP face recognition system for SRE' 19
197(11)
9.3.1 Survey of face recognition systems
197(3)
9.3.2 TSP face recognition system pipeline
200(1)
9.3.3 Databases used in the TSP face recognition system
201(1)
9.3.4 Face preprocessing
202(3)
9.3.5 Embedding extractor
205(1)
Initial version of the DNN architecture
205(1)
Final version of the DNN architecture
205(3)
9.3.6 Conclusions
208(1)
9.4 Audiovisual biometric system for the SRE' 19 multimedia challenge
208(4)
9.5 Conclusions and perspectives
212(7)
Acknowledgements
214(1)
References
214(5)
10 Voice biometrics: future trends and challenges ahead
219(4)
Douglas Reynolds
Craig S. Greenberg
10.1 Applications
219(1)
10.2 Privacy and security
220(1)
10.3 Research
221(2)
References 223(4)
Index 227
Carmen García-Mateo is a professor in the Department of Signal Theory and Communications of the University of Vigo (Spain) and the director of the Multimedia Research Group (GTM). Her research interests include Speech Technology, Audio Segmentation and Biometrics. She was the recipient of the '2014 Xunta de Galicia Josefa Wonenburger Award' for her outstanding career in the fields of science and technology. She received her PhD Degree from the Technical University of Madrid, Spain.



Gérard Chollet is VP of Research at Intelligent Voice, UK. His main research interests include Phonetics, Automatic Audio-Visual Speech Processing, Spoken Dialog Systems, Multimedia, Pattern Recognition, Biometrics, Privacy-Preserving Digital Signal Processing, Speech Pathology and Speech Training Aids. In 1983, he joined a newly created CNRS research unit at ENST (Telecom-ParisTech within the Institut Mines-Telecom). In 1992 he was asked to participate in the development of IDIAP, a new research laboratory of the Fondation Dalle Molle in Martigny, Switzerland. In July 2012 the CNRS granted him an emeritus status. He holds a PhD Degree in Computer Science and Linguistics from the University of California, Santa Barbara, USA.
Lisainfo e-raamatute kohta Lisainfo e-raamatute kohta
Püsilink: https://www.kriso.ee/db/97817856190146e.html
Märksõnad: