SRE08 system. Nir Krause Ran Gazit Gennady Karvitsky. Leave Impersonators, fraudsters and identity thieves speechless

Size: px
Start display at page:

Download "SRE08 system. Nir Krause Ran Gazit Gennady Karvitsky. Leave Impersonators, fraudsters and identity thieves speechless"

Transcription

1 Leave Impersonators, fraudsters and identity thieves speechless SRE08 system Nir Krause Ran Gazit Gennady Karvitsky Copyright 2008 PerSay Inc. All Rights Reserved

2 Focus: Multilingual telephone speech and 10sec conditions

3

4 Qualcomm-ICSI- OGI Wiener filter (MIC only)

5 MFCC & LPCC

6 SGM Svm in the Gmm Models space SGM= GMM-SVM=GSV=GMS=?

7 NAP SGM

8 GMM

9 TNO Thank you David!

10 Tuning with Focal Thank you Niko!

11 Condition Short2-short3 PRS1 (Primary) LPCC NAP SGM + MFCC NAP SGM + TNO Short2-10sec LPCC NAP SGM + MFCC NAP SGM 10sec-10sec LPCC SGM + MFCC SGM + GMM Short2-summed LPCC NAP SGM + MFCC NAP SGM

12

13 Super vectors

14 Super vector generation m 1 :Means-only Bayesian adaptation -> UBM with 512 Gaussians (m ubm ) top 10 scoring Gaussians A super vector of Gaussian means m 2 =m 1 -m ubm m ( gaussian, feature ) m 3 2 m=m 3 / m 3 L2 normalization. ( gaussian, feature ) weight ( gaussian ) var( gaussian, feature )

15 Data engineering

16 Strategy: Gender dependent Sub-condition dependent Greedy

17

18 Optimized parameters UBM, Negative & NAP speakers (among a few choices) NAP dimension Relevance factor

19 LPCC & MFCC NAP SGM short2-short3 phone (in both train & test), short2-summed Males Females Description UBM background data segments of different speakers from SREs: 99, 03, 04 & 05. Negative examples segments from Call Friend, and SREs: 99, 03, 04 & 05. NAP different speakers with at least 6 calls in SREs 04 & 05, who do not appear in the negative examples. NAP dimension Relevance factor 3 3

20 LPCC NAP SGM short2-short3 microphone (in either test or train) Males Females Description UBM background data segments of different speakers from SREs: 99, 03, 04 & 05. Negative examples segments from SREs: 99, 03, 04 & 05, including 05 mic tests data. NAP different speakers with at least 6 calls in SREs 04 & 05mic, who do not appear in the negative examples. NAP dimension Relevance factor 3 3

21 LPCC NAP SGM short2-10sec (MFCC used slightly different background data) Males Females description UBM background data segments of different speakers from SREs: 99, 03, 04 & 05. Negative examples segments from SREs: 99, 03, 04 & 05. Only the first 15sec of net audio were used to create the super vector, to match the test segment length. NAP different speakers with at least 6 calls in SREs 04 & 05, who do not appear in the negative examples. NAP dimension Relevance factor 3 3

22 LPCC & MFCC SGM 10sec-10sec Males Females description UBM background data segments of different speakers from SREs: 99, 03, 04 & 05. Negative examples segments from SREs: 99, 03, 04 & 05. Only the first 15sec of net audio were used to create the super vector, to match the test segment length. Relevance factor 1 1 Other The silence detector parameters were optimized for this condition, to extract more frames

23 10sec-10sec GMM Same as last year

24

25 Equal fusion Focal logistic regression

26 Good old SRE06 (didn t use the new short-short lists)

27 Results

28 Short2-short3 Int-Int Int-Int same PRS1 (NAP SGM + TNO) Int-Int different Int-Tel Tel-Mic Tel-Tel Tel-Tel Eng EER Tel-Tel native Eng mindcf actdcf PRS2 (NAP SGM) EER mindcf actdcf

29 Short2-summed Tel-tel Tel-tel Eng EER mindcf actdcf

30 Short2-10sec Tel-tel Tel-tel Eng EER mindcf actdcf

31 10sec-10sec Tel-tel Tel-tel Eng PRS1 EER mindcf actdcf PRS2 (SGM) EER mindcf actdcf PRS3 (Tnormed GMM) EER mindcf actdcf

32 Road to no-where Z/T/ZT norm for SGM? Only in GMM

33 Road to no-where 1024 Gaussians

34 Road to no-where Concatenate a male adapted & a female adapted super vectors (as in SRI s MLLR)

35 Road to no-where Wiener filter on telephone data

36 Road to no-where Factor analysis session compensation: as good as NAP, doesn t improve much with fusion

37 Road to no-where NAP on 10sec training duration doesn t help. Helps when training on 2.5 min, testing 10sec

38 Road to no-where Fusion of same system with different background datausually not useful

39 Back to the future Joint Factor Analysis still in process

40 PerSay s NIST VS customers (mainly call centers of banks, telecoms etc...)

41 Dev & Test data NIST Dev: SRE 06,05,04, Different background and development data Thousands of speakers SRE target models 100,000 tests 100GB Customers Same dev & test: speakers 10 speakers 3 speakers

42 Duration NIST Focus on Train: 2.5 minutes Test: 2.5 minutes Customers Train: ~1 minute Test: 20 sec summed Can we remove the agent?

43 TI/TD NIST Text Independent Customers Text Dependent (90%) 0.5-3% EER Text Independent (10%)

44 Thank You Copyright 2008 PerSay Inc. All Rights Reserved

Comparative Evaluation of Feature Normalization Techniques for Speaker Verification

Comparative Evaluation of Feature Normalization Techniques for Speaker Verification Comparative Evaluation of Feature Normalization Techniques for Speaker Verification Md Jahangir Alam 1,2, Pierre Ouellet 1, Patrick Kenny 1, Douglas O Shaughnessy 2, 1 CRIM, Montreal, Canada {Janagir.Alam,

More information

ABC submission for NIST SRE 2016

ABC submission for NIST SRE 2016 ABC submission for NIST SRE 2016 Agnitio+BUT+CRIM Oldrich Plchot, Pavel Matejka, Ondrej Novotny, Anna Silnova, Johan Rohdin, Mireia Diez, Ondrej Glembek, Xiaowei Jiang, Lukas Burget, Martin Karafiat, Lucas

More information

arxiv: v1 [cs.sd] 8 Jun 2017

arxiv: v1 [cs.sd] 8 Jun 2017 SUT SYSTEM DESCRIPTION FOR NIST SRE 2016 Hossein Zeinali 1,2, Hossein Sameti 1 and Nooshin Maghsoodi 1 1 Sharif University of Technology, Tehran, Iran 2 Brno University of Technology, Speech@FIT and IT4I

More information

SUT Submission for NIST 2016 Speaker Recognition Evaluation: Description and Analysis

SUT Submission for NIST 2016 Speaker Recognition Evaluation: Description and Analysis The 2017 Conference on Computational Linguistics and Speech Processing ROCLING 2017, pp. 276-286 The Association for Computational Linguistics and Chinese Language Processing SUT Submission for NIST 2016

More information

Vulnerability of Voice Verification System with STC anti-spoofing detector to different methods of spoofing attacks

Vulnerability of Voice Verification System with STC anti-spoofing detector to different methods of spoofing attacks Vulnerability of Voice Verification System with STC anti-spoofing detector to different methods of spoofing attacks Vadim Shchemelinin 1,2, Alexandr Kozlov 2, Galina Lavrentyeva 2, Sergey Novoselov 1,2

More information

PDF hosted at the Radboud Repository of the Radboud University Nijmegen

PDF hosted at the Radboud Repository of the Radboud University Nijmegen PDF hosted at the Radboud Repository of the Radboud University Nijmegen The following full text is a publisher's version. For additional information about this publication click this link. http://hdl.handle.net/2066/94752

More information

FOUR WEIGHTINGS AND A FUSION: A CEPSTRAL-SVM SYSTEM FOR SPEAKER RECOGNITION. Sachin S. Kajarekar

FOUR WEIGHTINGS AND A FUSION: A CEPSTRAL-SVM SYSTEM FOR SPEAKER RECOGNITION. Sachin S. Kajarekar FOUR WEIGHTINGS AND A FUSION: A CEPSTRAL-SVM SYSTEM FOR SPEAKER RECOGNITION Sachin S. Kajarekar Speech Technology and Research Laboratory SRI International, Menlo Park, CA, USA sachin@speech.sri.com ABSTRACT

More information

Bo#leneck Features from SNR- Adap9ve Denoising Deep Classifier for Speaker Iden9fica9on

Bo#leneck Features from SNR- Adap9ve Denoising Deep Classifier for Speaker Iden9fica9on Bo#leneck Features from SNR- Adap9ve Denoising Deep Classifier for Speaker Iden9fica9on TAN Zhili & MAK Man-Wai APSIPA 2015 Department of Electronic and Informa2on Engineering The Hong Kong Polytechnic

More information

SAS: A speaker verification spoofing database containing diverse attacks

SAS: A speaker verification spoofing database containing diverse attacks SAS: A speaker verification spoofing database containing diverse attacks Zhizheng Wu 1, Ali Khodabakhsh 2, Cenk Demiroglu 2, Junichi Yamagishi 1,3, Daisuke Saito 4, Tomoki Toda 5, Simon King 1 1 University

More information

A ROBUST SPEAKER CLUSTERING ALGORITHM

A ROBUST SPEAKER CLUSTERING ALGORITHM A ROBUST SPEAKER CLUSTERING ALGORITHM J. Ajmera IDIAP P.O. Box 592 CH-1920 Martigny, Switzerland jitendra@idiap.ch C. Wooters ICSI 1947 Center St., Suite 600 Berkeley, CA 94704, USA wooters@icsi.berkeley.edu

More information

THE 2013 SPEAKER RECOGNITION EVALUATION IN MOBILE ENVIRONMENT

THE 2013 SPEAKER RECOGNITION EVALUATION IN MOBILE ENVIRONMENT RESEARCH IDIAP REPORT THE 013 SPEAKER RECOGNITION EVALUATION IN MOBILE ENVIRONMENT Elie Khoury Bostjan Vesnicer Javier Franco-Pedroso Ricardo Violato Zenelabidine Boulkenafet Luis-Miguel Mazaira Fernandez

More information

Multifactor Fusion for Audio-Visual Speaker Recognition

Multifactor Fusion for Audio-Visual Speaker Recognition Proceedings of the 7th WSEAS International Conference on Signal, Speech and Image Processing, Beijing, China, September 15-17, 2007 70 Multifactor Fusion for Audio-Visual Speaker Recognition GIRIJA CHETTY

More information

Voiceprint-based Access Control for Wireless Insulin Pump Systems

Voiceprint-based Access Control for Wireless Insulin Pump Systems Voiceprint-based Access Control for Wireless Insulin Pump Systems Presenter: Xiaojiang Du Bin Hao, Xiali Hei, Yazhou Tu, Xiaojiang Du, and Jie Wu School of Computing and Informatics, University of Louisiana

More information

Multimedia Event Detection for Large Scale Video. Benjamin Elizalde

Multimedia Event Detection for Large Scale Video. Benjamin Elizalde Multimedia Event Detection for Large Scale Video Benjamin Elizalde Outline Motivation TrecVID task Related work Our approach (System, TF/IDF) Results & Processing time Conclusion & Future work Agenda 2

More information

Trial-Based Calibration for Speaker Recognition in Unseen Conditions

Trial-Based Calibration for Speaker Recognition in Unseen Conditions Trial-Based Calibration for Speaker Recognition in Unseen Conditions Mitchell McLaren, Aaron Lawson, Luciana Ferrer, Nicolas Scheffer, Yun Lei Speech Technology and Research Laboratory SRI International,

More information

Hands On: Multimedia Methods for Large Scale Video Analysis (Lecture) Dr. Gerald Friedland,

Hands On: Multimedia Methods for Large Scale Video Analysis (Lecture) Dr. Gerald Friedland, Hands On: Multimedia Methods for Large Scale Video Analysis (Lecture) Dr. Gerald Friedland, fractor@icsi.berkeley.edu 1 Today Recap: Some more Machine Learning Multimedia Systems An example Multimedia

More information

The Approach of Mean Shift based Cosine Dissimilarity for Multi-Recording Speaker Clustering

The Approach of Mean Shift based Cosine Dissimilarity for Multi-Recording Speaker Clustering The Approach of Mean Shift based Cosine Dissimilarity for Multi-Recording Speaker Clustering 1 D. Jareena Begum, 2 K Rajendra Prasad, 3 M Suleman Basha 1 M.Tech in SE, RGMCET, Nandyal 2 Assoc Prof, Dept

More information

Manual operations of the voice identification program GritTec's Speaker-ID: The Mobile Client

Manual operations of the voice identification program GritTec's Speaker-ID: The Mobile Client Manual operations of the voice identification program GritTec's Speaker-ID: The Mobile Client Version 4.00 2017 Title Short name of product Version 4.00 Manual operations of GritTec s Speaker-ID: The Mobile

More information

Introducing I-Vectors for Joint Anti-spoofing and Speaker Verification

Introducing I-Vectors for Joint Anti-spoofing and Speaker Verification Introducing I-Vectors for Joint Anti-spoofing and Speaker Verification Elie Khoury, Tomi Kinnunen, Aleksandr Sizov, Zhizheng Wu, Sébastien Marcel Idiap Research Institute, Switzerland School of Computing,

More information

A Low-Complexity Dynamic Face-Voice Feature Fusion Approach to Multimodal Person Recognition

A Low-Complexity Dynamic Face-Voice Feature Fusion Approach to Multimodal Person Recognition A Low-Complexity Dynamic Face-Voice Feature Fusion Approach to Multimodal Person Recognition Dhaval Shah, Kyu J. Han, Shrikanth S. Nayaranan Ming Hsieh Department of Electrical Engineering, Viterbi School

More information

Pitch Prediction from Mel-frequency Cepstral Coefficients Using Sparse Spectrum Recovery

Pitch Prediction from Mel-frequency Cepstral Coefficients Using Sparse Spectrum Recovery Pitch Prediction from Mel-frequency Cepstral Coefficients Using Sparse Spectrum Recovery Achuth Rao MV, Prasanta Kumar Ghosh SPIRE LAB Electrical Engineering, Indian Institute of Science (IISc), Bangalore,

More information

Speaker Diarization System Based on GMM and BIC

Speaker Diarization System Based on GMM and BIC Speaer Diarization System Based on GMM and BIC Tantan Liu 1, Xiaoxing Liu 1, Yonghong Yan 1 1 ThinIT Speech Lab, Institute of Acoustics, Chinese Academy of Sciences Beijing 100080 {tliu, xliu,yyan}@hccl.ioa.ac.cn

More information

Applications of Keyword-Constraining in Speaker Recognition. Howard Lei. July 2, Introduction 3

Applications of Keyword-Constraining in Speaker Recognition. Howard Lei. July 2, Introduction 3 Applications of Keyword-Constraining in Speaker Recognition Howard Lei hlei@icsi.berkeley.edu July 2, 2007 Contents 1 Introduction 3 2 The keyword HMM system 4 2.1 Background keyword HMM training............................

More information

IMPROVED SPEAKER RECOGNITION USING DCT COEFFICIENTS AS FEATURES. Mitchell McLaren, Yun Lei

IMPROVED SPEAKER RECOGNITION USING DCT COEFFICIENTS AS FEATURES. Mitchell McLaren, Yun Lei IMPROVED SPEAKER RECOGNITION USING DCT COEFFICIENTS AS FEATURES Mitchell McLaren, Yun Lei Speech Technology and Research Laboratory, SRI International, California, USA {mitch,yunlei}@speech.sri.com ABSTRACT

More information

GYROPHONE RECOGNIZING SPEECH FROM GYROSCOPE SIGNALS. Yan Michalevsky (1), Gabi Nakibly (2) and Dan Boneh (1)

GYROPHONE RECOGNIZING SPEECH FROM GYROSCOPE SIGNALS. Yan Michalevsky (1), Gabi Nakibly (2) and Dan Boneh (1) GYROPHONE RECOGNIZING SPEECH FROM GYROSCOPE SIGNALS Yan Michalevsky (1), Gabi Nakibly (2) and Dan Boneh (1) (1) Stanford University (2) National Research and Simulation Center, Rafael Ltd. 0 MICROPHONE

More information

Comparison of Clustering Methods: a Case Study of Text-Independent Speaker Modeling

Comparison of Clustering Methods: a Case Study of Text-Independent Speaker Modeling Comparison of Clustering Methods: a Case Study of Text-Independent Speaker Modeling Tomi Kinnunen, Ilja Sidoroff, Marko Tuononen, Pasi Fränti Speech and Image Processing Unit, School of Computing, University

More information

Complex Identification Decision Based on Several Independent Speaker Recognition Methods. Ilya Oparin Speech Technology Center

Complex Identification Decision Based on Several Independent Speaker Recognition Methods. Ilya Oparin Speech Technology Center Complex Identification Decision Based on Several Independent Speaker Recognition Methods Ilya Oparin Speech Technology Center Corporate Overview Global provider of voice biometric solutions Company name:

More information

Speaker Verification with Adaptive Spectral Subband Centroids

Speaker Verification with Adaptive Spectral Subband Centroids Speaker Verification with Adaptive Spectral Subband Centroids Tomi Kinnunen 1, Bingjun Zhang 2, Jia Zhu 2, and Ye Wang 2 1 Speech and Dialogue Processing Lab Institution for Infocomm Research (I 2 R) 21

More information

Improving Speaker Verification Performance in Presence of Spoofing Attacks Using Out-of-Domain Spoofed Data

Improving Speaker Verification Performance in Presence of Spoofing Attacks Using Out-of-Domain Spoofed Data INTERSPEECH 17 August 24, 17, Stockholm, Sweden Improving Speaker Verification Performance in Presence of Spoofing Attacks Using Out-of-Domain Spoofed Data Achintya Kr. Sarkar 1, Md. Sahidullah 2, Zheng-Hua

More information

Lec 08 Feature Aggregation II: Fisher Vector, Super Vector and AKULA

Lec 08 Feature Aggregation II: Fisher Vector, Super Vector and AKULA Image Analysis & Retrieval CS/EE 5590 Special Topics (Class Ids: 44873, 44874) Fall 2016, M/W 4-5:15pm@Bloch 0012 Lec 08 Feature Aggregation II: Fisher Vector, Super Vector and AKULA Zhu Li Dept of CSEE,

More information

Neetha Das Prof. Andy Khong

Neetha Das Prof. Andy Khong Neetha Das Prof. Andy Khong Contents Introduction and aim Current system at IMI Proposed new classification model Support Vector Machines Initial audio data collection and processing Features and their

More information

Improving Robustness to Compressed Speech in Speaker Recognition

Improving Robustness to Compressed Speech in Speaker Recognition INTERSPEECH 2013 Improving Robustness to Compressed Speech in Speaker Recognition Mitchell McLaren 1, Victor Abrash 1, Martin Graciarena 1, Yun Lei 1, Jan Pe sán 2 1 Speech Technology and Research Laboratory,

More information

COMBINING FEATURE SETS WITH SUPPORT VECTOR MACHINES: APPLICATION TO SPEAKER RECOGNITION

COMBINING FEATURE SETS WITH SUPPORT VECTOR MACHINES: APPLICATION TO SPEAKER RECOGNITION COMBINING FEATURE SETS WITH SUPPORT VECTOR MACHINES: APPLICATION TO SPEAKER RECOGNITION Andrew O. Hatch ;2, Andreas Stolcke ;3, and Barbara Peskin The International Computer Science Institute, Berkeley,

More information

Towards PLDA-RBM based Speaker Recognition in Mobile Environment: Designing Stacked/Deep PLDA-RBM Systems

Towards PLDA-RBM based Speaker Recognition in Mobile Environment: Designing Stacked/Deep PLDA-RBM Systems Nautch, Hao, Stafylaki, Rathgeb, Buch PLDA-RBM mobile data / Shanghai, 23.03.2016 1/14 Toward PLDA-RBM baed Speaker Recognition in Mobile Environment: Deigning Stacked/Deep PLDA-RBM Sytem A. Nautch, H.

More information

The research on Uighur speaker-dependent isolated word speech recognition

The research on Uighur speaker-dependent isolated word speech recognition The research on Uighur speaker-dependent isolated word speech recognition Wushour silamu Caiqin Nuominghua College of information science and engineering Xinjiang University, Urumqi 830046 Abstract: A

More information

Project 3 Q&A. Jonathan Krause

Project 3 Q&A. Jonathan Krause Project 3 Q&A Jonathan Krause 1 Outline R-CNN Review Error metrics Code Overview Project 3 Report Project 3 Presentations 2 Outline R-CNN Review Error metrics Code Overview Project 3 Report Project 3 Presentations

More information

Voice. Voice. Patterson EagleSoft Overview Voice 629

Voice. Voice. Patterson EagleSoft Overview Voice 629 Voice Voice Using the Microsoft voice engine, Patterson EagleSoft's Voice module is now faster, easier and more efficient than ever. Please refer to your Voice Installation guide prior to installing the

More information

HANDSET-DEPENDENT BACKGROUND MODELS FOR ROBUST. Larry P. Heck and Mitchel Weintraub. Speech Technology and Research Laboratory.

HANDSET-DEPENDENT BACKGROUND MODELS FOR ROBUST. Larry P. Heck and Mitchel Weintraub. Speech Technology and Research Laboratory. HANDSET-DEPENDENT BACKGROUND MODELS FOR ROBUST TEXT-INDEPENDENT SPEAKER RECOGNITION Larry P. Heck and Mitchel Weintraub Speech Technology and Research Laboratory SRI International Menlo Park, CA 9 ABSTRACT

More information

Client Dependent GMM-SVM Models for Speaker Verification

Client Dependent GMM-SVM Models for Speaker Verification Client Dependent GMM-SVM Models for Speaker Verification Quan Le, Samy Bengio IDIAP, P.O. Box 592, CH-1920 Martigny, Switzerland {quan,bengio}@idiap.ch Abstract. Generative Gaussian Mixture Models (GMMs)

More information

SUPERVISED LEARNING METHODS. Stanley Liang, PhD Candidate, Lassonde School of Engineering, York University Helix Science Engagement Programs 2018

SUPERVISED LEARNING METHODS. Stanley Liang, PhD Candidate, Lassonde School of Engineering, York University Helix Science Engagement Programs 2018 SUPERVISED LEARNING METHODS Stanley Liang, PhD Candidate, Lassonde School of Engineering, York University Helix Science Engagement Programs 2018 2 CHOICE OF ML You cannot know which algorithm will work

More information

Xing Fan, Carlos Busso and John H.L. Hansen

Xing Fan, Carlos Busso and John H.L. Hansen Xing Fan, Carlos Busso and John H.L. Hansen Center for Robust Speech Systems (CRSS) Erik Jonsson School of Engineering & Computer Science Department of Electrical Engineering University of Texas at Dallas

More information

Variable Selection 6.783, Biomedical Decision Support

Variable Selection 6.783, Biomedical Decision Support 6.783, Biomedical Decision Support (lrosasco@mit.edu) Department of Brain and Cognitive Science- MIT November 2, 2009 About this class Why selecting variables Approaches to variable selection Sparsity-based

More information

Minimal-Impact Personal Audio Archives

Minimal-Impact Personal Audio Archives Minimal-Impact Personal Audio Archives Dan Ellis, Keansub Lee, Jim Ogle Laboratory for Recognition and Organization of Speech and Audio Dept. Electrical Eng., Columbia Univ., NY USA dpwe@ee.columbia.edu

More information

STC ANTI-SPOOFING SYSTEMS FOR THE ASVSPOOF 2015 CHALLENGE

STC ANTI-SPOOFING SYSTEMS FOR THE ASVSPOOF 2015 CHALLENGE STC ANTI-SPOOFING SYSTEMS FOR THE ASVSPOOF 2015 CHALLENGE Sergey Novoselov 1,2, Alexandr Kozlov 2, Galina Lavrentyeva 1,2, Konstantin Simonchik 1,2, Vadim Shchemelinin 1,2 1 ITMO University, St. Petersburg,

More information

Inter-session Variability Modelling and Joint Factor Analysis for Face Authentication

Inter-session Variability Modelling and Joint Factor Analysis for Face Authentication Inter-session Variability Modelling and Joint Factor Analysis for Face Authentication Roy Wallace Idiap Research Institute, Martigny, Switzerland roy.wallace@idiap.ch Mitchell McLaren Radboud University

More information

A text-independent speaker verification model: A comparative analysis

A text-independent speaker verification model: A comparative analysis A text-independent speaker verification model: A comparative analysis Rishi Charan, Manisha.A, Karthik.R, Raesh Kumar M, Senior IEEE Member School of Electronic Engineering VIT University Tamil Nadu, India

More information

Probabilistic scoring using decision trees for fast and scalable speaker recognition

Probabilistic scoring using decision trees for fast and scalable speaker recognition Probabilistic scoring using decision trees for fast and scalable speaker recognition Gilles Gonon, Frédéric Bimbot, Rémi Gribonval To cite this version: Gilles Gonon, Frédéric Bimbot, Rémi Gribonval. Probabilistic

More information

10601 Machine Learning. Model and feature selection

10601 Machine Learning. Model and feature selection 10601 Machine Learning Model and feature selection Model selection issues We have seen some of this before Selecting features (or basis functions) Logistic regression SVMs Selecting parameter value Prior

More information

CAPTURING AUDIO DATA FAQS

CAPTURING AUDIO DATA FAQS EVS AUDIO FAQ CAPTURING AUDIO DATA FAQS EVS is a powerful engine that turns audio into a rich data stream for use in upstream applications such as analytics or CRM systems. The better the quality of the

More information

The BioSecure Talking-Face Reference System

The BioSecure Talking-Face Reference System The BioSecure Talking-Face Reference System Hervé Bredin 1, Guido Aversano 1, Chafic Mokbel 2 and Gérard Chollet 1 1 CNRS-LTCI, GET-ENST (TSI Department), 46 rue Barrault, 75013 Paris, France 2 University

More information

babytel Self Install Guide

babytel Self Install Guide babytel Self Install Guide Last updated 4/5/2005 CONTENTS What you need ------------------------------------------------------------------------------------------------------------3 Download & Install

More information

Multi-modal Person Identification in a Smart Environment

Multi-modal Person Identification in a Smart Environment Multi-modal Person Identification in a Smart Environment Hazım Kemal Ekenel 1, Mika Fischer 1, Qin Jin 2, Rainer Stiefelhagen 1 1 Interactive Systems Labs (ISL), Universität Karlsruhe (TH), 76131 Karlsruhe,

More information

Passive Detection. What is KIVOX Passive Detection? Product Datasheet. Key Benefits. 3 APIs in one product

Passive Detection. What is KIVOX Passive Detection? Product Datasheet. Key Benefits. 3 APIs in one product Passive Detection Product Datasheet KIVOX Passive Detection is part of the KIVOX 4.0 family of products. KIVOX 4.0 brings our partners the full potential of real voice biometrics technology by providing

More information

ON THE EFFECT OF SCORE EQUALIZATION IN SVM MULTIMODAL BIOMETRIC SYSTEMS

ON THE EFFECT OF SCORE EQUALIZATION IN SVM MULTIMODAL BIOMETRIC SYSTEMS ON THE EFFECT OF SCORE EQUALIZATION IN SVM MULTIMODAL BIOMETRIC SYSTEMS Pascual Ejarque and Javier Hernando TALP Research Center, Department of Signal Theory and Communications Technical University of

More information

Gender-dependent acoustic models fusion developed for automatic subtitling of Parliament meetings broadcasted by the Czech TV

Gender-dependent acoustic models fusion developed for automatic subtitling of Parliament meetings broadcasted by the Czech TV Gender-dependent acoustic models fusion developed for automatic subtitling of Parliament meetings broadcasted by the Czech TV Jan Vaněk and Josef V. Psutka Department of Cybernetics, West Bohemia University,

More information

How accurate is AGNITIO KIVOX Voice ID?

How accurate is AGNITIO KIVOX Voice ID? How accurate is AGNITIO KIVOX Voice ID? Overview Using natural speech, KIVOX can work with error rates below 1%. When optimized for short utterances, where the same phrase is used for enrolment and authentication,

More information

MACHINE LEARNING Example: Google search

MACHINE LEARNING Example: Google search MACHINE LEARNING Lauri Ilison, PhD Data Scientist 20.11.2014 Example: Google search 1 27.11.14 Facebook: 350 million photo uploads every day The dream is to build full knowledge of the world and know everything

More information

Joint Optimisation of Tandem Systems using Gaussian Mixture Density Neural Network Discriminative Sequence Training

Joint Optimisation of Tandem Systems using Gaussian Mixture Density Neural Network Discriminative Sequence Training Joint Optimisation of Tandem Systems using Gaussian Mixture Density Neural Network Discriminative Sequence Training Chao Zhang and Phil Woodland March 8, 07 Cambridge University Engineering Department

More information

Multi-Modal Human- Computer Interaction

Multi-Modal Human- Computer Interaction Multi-Modal Human- Computer Interaction Attila Fazekas University of Debrecen, Hungary Road Map Multi-modal interactions and systems (main categories, examples, benefits) Face detection, facial gestures

More information

Dr Andrew Abel University of Stirling, Scotland

Dr Andrew Abel University of Stirling, Scotland Dr Andrew Abel University of Stirling, Scotland University of Stirling - Scotland Cognitive Signal Image and Control Processing Research (COSIPRA) Cognitive Computation neurobiology, cognitive psychology

More information

MACHINE LEARNING: CLUSTERING, AND CLASSIFICATION. Steve Tjoa June 25, 2014

MACHINE LEARNING: CLUSTERING, AND CLASSIFICATION. Steve Tjoa June 25, 2014 MACHINE LEARNING: CLUSTERING, AND CLASSIFICATION Steve Tjoa kiemyang@gmail.com June 25, 2014 Review from Day 2 Supervised vs. Unsupervised Unsupervised - clustering Supervised binary classifiers (2 classes)

More information

TA Section: Problem Set 4

TA Section: Problem Set 4 TA Section: Problem Set 4 Outline Discriminative vs. Generative Classifiers Image representation and recognition models Bag of Words Model Part-based Model Constellation Model Pictorial Structures Model

More information

Operating Instructions

Operating Instructions IBT5 07/11/2007 ECR 2014 BT5 Operating Instructions Baldwin Boxall Communications Ltd. Wealden Industrial Estate, Farningham Road Crowborough, East Sussex, TN6 2JR Telephone: 01892 664422 Fax: 01892 663146

More information

k-nearest Neighbor (knn) Sept Youn-Hee Han

k-nearest Neighbor (knn) Sept Youn-Hee Han k-nearest Neighbor (knn) Sept. 2015 Youn-Hee Han http://link.koreatech.ac.kr ²Eager Learners Eager vs. Lazy Learning when given a set of training data, it will construct a generalization model before receiving

More information

SPEECH FEATURE EXTRACTION USING WEIGHTED HIGHER-ORDER LOCAL AUTO-CORRELATION

SPEECH FEATURE EXTRACTION USING WEIGHTED HIGHER-ORDER LOCAL AUTO-CORRELATION Far East Journal of Electronics and Communications Volume 3, Number 2, 2009, Pages 125-140 Published Online: September 14, 2009 This paper is available online at http://www.pphmj.com 2009 Pushpa Publishing

More information

DUET Operate Instruction

DUET Operate Instruction DUET Operate Instruction Version1.0.0 INDEX 1.INTRODUCTION 2 2.APPEARANCE 3 3.CONTENT 4 4.SYSTEM REQUIREMENT. 4 5.INSTALLATION 5 6.MAIN MENU..7 7.SETUP. 8 8.RECORD. 12 9.SERVICE AND WARRANTY.18 10.CONTACT

More information

Self Lane Assignment Using Smart Mobile Camera For Intelligent GPS Navigation and Traffic Interpretation

Self Lane Assignment Using Smart Mobile Camera For Intelligent GPS Navigation and Traffic Interpretation For Intelligent GPS Navigation and Traffic Interpretation Tianshi Gao Stanford University tianshig@stanford.edu 1. Introduction Imagine that you are driving on the highway at 70 mph and trying to figure

More information

Analyzing Vocal Patterns to Determine Emotion Maisy Wieman, Andy Sun

Analyzing Vocal Patterns to Determine Emotion Maisy Wieman, Andy Sun Analyzing Vocal Patterns to Determine Emotion Maisy Wieman, Andy Sun 1. Introduction The human voice is very versatile and carries a multitude of emotions. Emotion in speech carries extra insight about

More information

Rossmann Store Sales. 1 Introduction. 3 Datasets and Features. 2 Related Work. David Beam and Mark Schramm. December 2015

Rossmann Store Sales. 1 Introduction. 3 Datasets and Features. 2 Related Work. David Beam and Mark Schramm. December 2015 Rossmann Store Sales David Beam and Mark Schramm 1 Introduction December 015 The objective of this project is to forecast sales in euros at 1115 stores owned by Rossmann, a European pharmaceutical company.

More information

Hands On: Multimedia Methods for Large Scale Video Analysis (Lecture) Dr. Gerald Friedland,

Hands On: Multimedia Methods for Large Scale Video Analysis (Lecture) Dr. Gerald Friedland, Hands On: Multimedia Methods for Large Scale Video Analysis (Lecture) Dr. Gerald Friedland, fractor@icsi.berkeley.edu 1 Today Answers to Questions How to estimate resources for large data projects - Some

More information

Excel 2007/2010/2013: Using Data Validation to provide dropdown selection menu

Excel 2007/2010/2013: Using Data Validation to provide dropdown selection menu Excel 2007/2010/2013: Using Data Validation to provide dropdown selection menu Submitted by Jess on Sun, 06/30/2013-20:57 In Excel, there are various ways to provide a drop-down menu in a form or in cells.

More information

Discriminative training and Feature combination

Discriminative training and Feature combination Discriminative training and Feature combination Steve Renals Automatic Speech Recognition ASR Lecture 13 16 March 2009 Steve Renals Discriminative training and Feature combination 1 Overview Hot topics

More information

Confidence Measures: how much we can trust our speech recognizers

Confidence Measures: how much we can trust our speech recognizers Confidence Measures: how much we can trust our speech recognizers Prof. Hui Jiang Department of Computer Science York University, Toronto, Ontario, Canada Email: hj@cs.yorku.ca Outline Speech recognition

More information

Lecture 13: Model selection and regularization

Lecture 13: Model selection and regularization Lecture 13: Model selection and regularization Reading: Sections 6.1-6.2.1 STATS 202: Data mining and analysis October 23, 2017 1 / 17 What do we know so far In linear regression, adding predictors always

More information

Introduction to Deep Learning in Signal Processing & Communications with MATLAB

Introduction to Deep Learning in Signal Processing & Communications with MATLAB Introduction to Deep Learning in Signal Processing & Communications with MATLAB Dr. Amod Anandkumar Pallavi Kar Application Engineering Group, Mathworks India 2019 The MathWorks, Inc. 1 Different Types

More information

Deep Convolutional Neural Network using Triplet of Faces, Deep Ensemble, and Scorelevel Fusion for Face Recognition

Deep Convolutional Neural Network using Triplet of Faces, Deep Ensemble, and Scorelevel Fusion for Face Recognition IEEE 2017 Conference on Computer Vision and Pattern Recognition Deep Convolutional Neural Network using Triplet of Faces, Deep Ensemble, and Scorelevel Fusion for Face Recognition Bong-Nam Kang*, Yonghyun

More information

Detector. Flash. Detector

Detector. Flash. Detector CLIPS at TRECvid: Shot Boundary Detection and Feature Detection Georges M. Quénot, Daniel Moraru, and Laurent Besacier CLIPS-IMAG, BP53, 38041 Grenoble Cedex 9, France Georges.Quenot@imag.fr Abstract This

More information

LIA_SPKDET. Package documentation

LIA_SPKDET. Package documentation LIA_SPKDET Package documentation Edité par Eric Charton/ LIA Projet Mistral mistral.univ-avignon.fr 1/11 Crédits This documentation is a user guide: Editeur / Editor: Eric Charton eric.charton@univ-avignon.fr

More information

MULTIMODAL PERSON IDENTIFICATION IN A SMART ROOM. J.Luque, R.Morros, J.Anguita, M.Farrus, D.Macho, F.Marqués, C.Martínez, V.Vilaplana, J.

MULTIMODAL PERSON IDENTIFICATION IN A SMART ROOM. J.Luque, R.Morros, J.Anguita, M.Farrus, D.Macho, F.Marqués, C.Martínez, V.Vilaplana, J. MULTIMODAL PERSON IDENTIFICATION IN A SMART ROOM JLuque, RMorros, JAnguita, MFarrus, DMacho, FMarqués, CMartínez, VVilaplana, J Hernando Technical University of Catalonia (UPC) Jordi Girona, 1-3 D5, 08034

More information

Setting UP the UMI-1 with True RTA

Setting UP the UMI-1 with True RTA Setting UP the UMI-1 with True RTA PC based test gear or single purpose device? Is there an easier way? Sure. There are lots of options for a single purpose device that will allow you to make frequency

More information

VentriLock: Exploring voice-based authentication systems

VentriLock: Exploring voice-based authentication systems VentriLock: Exploring voice-based authentication systems Chaouki KASMI & José LOPES ESTEVES ANSSI, FRANCE Hack In Paris 06/2017 2 WHO WE ARE Chaouki Kasmi and José Lopes Esteves ANSSI-FNISA / Wireless

More information

The SecurePhone PDA Database, Experimental Protocol and Automatic Test Procedure for Multimodal User Authentication

The SecurePhone PDA Database, Experimental Protocol and Automatic Test Procedure for Multimodal User Authentication The SecurePhone PDA Database, Experimental Protocol and Automatic Test Procedure for Multimodal User Authentication A.C. Morris 1, J. Koreman 1, H. Sellahewa 2, J. Ehlers 2, S. Jassim 2, L. Allano 3, S.

More information

Agatha: Multimodal Biometric Authentication Platform in Large-Scale Databases

Agatha: Multimodal Biometric Authentication Platform in Large-Scale Databases Agatha: Multimodal Biometric Authentication Platform in Large-Scale Databases David Hernando David Gómez Javier Rodríguez Saeta Pascual Ejarque 2 Javier Hernando 2 Biometric Technologies S.L., Barcelona,

More information

Introduction to Stata Toy Program #1 Basic Descriptives

Introduction to Stata Toy Program #1 Basic Descriptives Introduction to Stata 2018-19 Toy Program #1 Basic Descriptives Summary The goal of this toy program is to get you in and out of a Stata session and, along the way, produce some descriptive statistics.

More information

Speech Recognition. Project: Phone Recognition using Sphinx. Chia-Ho Ling. Sunya Santananchai. Professor: Dr. Kepuska

Speech Recognition. Project: Phone Recognition using Sphinx. Chia-Ho Ling. Sunya Santananchai. Professor: Dr. Kepuska Speech Recognition Project: Phone Recognition using Sphinx Chia-Ho Ling Sunya Santananchai Professor: Dr. Kepuska Objective Use speech data corpora to build a model using CMU Sphinx.Apply a built model

More information

Input speech signal. Selected /Rejected. Pre-processing Feature extraction Matching algorithm. Database. Figure 1: Process flow in ASR

Input speech signal. Selected /Rejected. Pre-processing Feature extraction Matching algorithm. Database. Figure 1: Process flow in ASR Volume 5, Issue 1, January 2015 ISSN: 2277 128X International Journal of Advanced Research in Computer Science and Software Engineering Research Paper Available online at: www.ijarcsse.com Feature Extraction

More information

Presentation attack detection in voice biometrics

Presentation attack detection in voice biometrics Chapter 1 Presentation attack detection in voice biometrics Pavel Korshunov and Sébastien Marcel Idiap Research Institute, Martigny, Switzerland {pavel.korshunov,sebastien.marcel}@idiap.ch Recent years

More information

Scott Shaobing Chen & P.S. Gopalakrishnan. IBM T.J. Watson Research Center. as follows:

Scott Shaobing Chen & P.S. Gopalakrishnan. IBM T.J. Watson Research Center.   as follows: SPEAKER, ENVIRONMENT AND CHANNEL CHANGE DETECTION AND CLUSTERING VIA THE BAYESIAN INFORMATION CRITERION Scott Shaobing Chen & P.S. Gopalakrishnan IBM T.J. Watson Research Center email: schen@watson.ibm.com

More information

DS Machine Learning and Data Mining I. Alina Oprea Associate Professor, CCIS Northeastern University

DS Machine Learning and Data Mining I. Alina Oprea Associate Professor, CCIS Northeastern University DS 4400 Machine Learning and Data Mining I Alina Oprea Associate Professor, CCIS Northeastern University January 24 2019 Logistics HW 1 is due on Friday 01/25 Project proposal: due Feb 21 1 page description

More information

Yiqi Yan. May 10, 2017

Yiqi Yan. May 10, 2017 Yiqi Yan May 10, 2017 P a r t I F u n d a m e n t a l B a c k g r o u n d s Convolution Single Filter Multiple Filters 3 Convolution: case study, 2 filters 4 Convolution: receptive field receptive field

More information

TYPES OF VARIABLES, STRUCTURE OF DATASETS, AND BASIC STATA LAYOUT

TYPES OF VARIABLES, STRUCTURE OF DATASETS, AND BASIC STATA LAYOUT PRIMER FOR ACS OUTCOMES RESEARCH COURSE: TYPES OF VARIABLES, STRUCTURE OF DATASETS, AND BASIC STATA LAYOUT STEP 1: Install STATA statistical software. STEP 2: Read through this primer and complete the

More information

Dynamic Time Warping

Dynamic Time Warping Centre for Vision Speech & Signal Processing University of Surrey, Guildford GU2 7XH. Dynamic Time Warping Dr Philip Jackson Acoustic features Distance measures Pattern matching Distortion penalties DTW

More information

The LENA Advanced Data Extractor (ADEX) User Guide Version 1.1.2

The LENA Advanced Data Extractor (ADEX) User Guide Version 1.1.2 The LENA Advanced Data Extractor (ADEX) User Guide Version 1.1.2 ADEXUG20110602 Copyright 2011 LENA Foundation The LENA Advanced Data Extractor User Guide ii The LENA Advanced Data Extractor (ADEX) User

More information

ABSTRACT AUTOMATIC SPEECH CODEC IDENTIFICATION WITH APPLICATIONS TO TAMPERING DETECTION OF SPEECH RECORDINGS

ABSTRACT AUTOMATIC SPEECH CODEC IDENTIFICATION WITH APPLICATIONS TO TAMPERING DETECTION OF SPEECH RECORDINGS ABSTRACT Title of thesis: AUTOMATIC SPEECH CODEC IDENTIFICATION WITH APPLICATIONS TO TAMPERING DETECTION OF SPEECH RECORDINGS Jingting Zhou, Master of Engineering, 212 Thesis directed by: Professor Carol

More information

Optimizing feature representation for speaker diarization using PCA and LDA

Optimizing feature representation for speaker diarization using PCA and LDA Optimizing feature representation for speaker diarization using PCA and LDA itsikv@netvision.net.il Jean-Francois Bonastre jean-francois.bonastre@univ-avignon.fr Outline Speaker Diarization what is it?

More information

Generic SIP Interface Configuration Guide

Generic SIP Interface Configuration Guide Generic SIP Interface Configuration Guide ASL Document Ref.: U-0701-1497.docx Issue: 1 complete, approved - Date: 28/11/16 Part Number: M0664_TBD Contents 1 Introduction... 3 2 Configuration... 4 Additional

More information

Version 2.6. SurVo Advanced User s Guide

Version 2.6. SurVo Advanced User s Guide Version 2.6 SurVo Advanced User s Guide Contents What is a SurVo?...3 SurVo: Voice Survey Form...3 About the Documentation...3 Ifbyphone on the Web...3 Setting up a SurVo...4 Speech/Recording Options...4

More information

The Expected Performance Curve: a New Assessment Measure for Person Authentication

The Expected Performance Curve: a New Assessment Measure for Person Authentication The Expected Performance Curve: a New Assessment Measure for Person Authentication Samy Bengio Johnny Mariéthoz IDIAP CP 592, rue du Simplon4 192 Martigny, Switzerland {bengio,marietho}@idiap.ch Abstract

More information

Voice Quality Assessment for Mobile to SIP Call over Live 3G Network

Voice Quality Assessment for Mobile to SIP Call over Live 3G Network Abstract 132 Voice Quality Assessment for Mobile to SIP Call over Live 3G Network G.Venkatakrishnan, I-H.Mkwawa and L.Sun Signal Processing and Multimedia Communications, University of Plymouth, Plymouth,

More information

Selection Control Structure CSC128: FUNDAMENTALS OF COMPUTER PROBLEM SOLVING

Selection Control Structure CSC128: FUNDAMENTALS OF COMPUTER PROBLEM SOLVING Selection Control Structure CSC128: FUNDAMENTALS OF COMPUTER PROBLEM SOLVING MULTIPLE SELECTION To solve a problem that has several selection, use either of the following method: Multiple selection nested

More information