Session Variability Compensation in Automatic Speaker Recognition

Size: px
Start display at page:

Download "Session Variability Compensation in Automatic Speaker Recognition"

Transcription

1 Session Variability Compensation in Automatic Speaker Recognition Javier González Domínguez VII Jornadas MAVIR Universidad Autónoma de Madrid November 2012

2 Outline 1. The Inter-session Variability Problem 2. From Eigenfaces to Joint Factor Analysis 3. Factor Analysis in Speaker and Language Recognition: I. Theory II. Where and How III. Efficiency 4. Total Variability 5. PLDA 6. Results (NIST SRE10, SRE 12) 2/31

3 Outline 1. The Inter-session Variability Problem 2. From Eigenfaces to Joint Factor Analysis 3. Factor Analysis in Speaker and Language Recognition: I. Theory II. Where and How III. Efficiency 4. Total Variability 5. PLDA 6. Results (NIST SRE10, SRE 12) 3/31

4 The Inter-Session Variability Problem: Causes Inter-session variability: All phenomena causing two recordings of a same identity to be different. v Transmission channel effects (GSM, landline,...). v Transducer characteristics (microphone type,...). v Environment Noise (traffic, people speaking,...) v Intra-speaker variability (age, illness, emotions,...) 4/31

5 Factor Analysis: Basis Principles: 1. Variability as a continuous source rather than discrete. 2. Modeling both session and inter-speaker/language variability. Assumption: 1. Variability lies in a lower-dimensional subspace. 5/31

6 Outline 1. The Inter-session Variability Problem 2. From Eigenfaces to Joint Factor Analysis 3. Factor Analysis in Speaker and Language Recognition: I. Theory II. Where and How III. Efficiency 4. Total Variability 5. PLDA 6. Results (NIST SRE10, SRE 12) 6/31

7 Eigenfaces, working scheme DEVELOPMENT STAGE Three first principal components Dev data: m samples Coding B PCA(C) DxM A Visualization DxK Train data: t samples TRAIN STAGE Reconstructed training images Coding M DxT A T M M KxT Visualization Test sample TEST STAGE s i! " accepted Coding t Dx1 A T M t Kx1 d(t,m) S Tx1 s i > " rejected 7/31

8 The GMM-UBM Framework: Maximum a Posteriori (MAP) cj cj x x x x x x x x x x x x x UBM ci Speaker Model A B ci 8/31

9 The GMM-UBM Framework: The Supervector Concept cj UBM ci DIMENSIONALITY! UBM = {" i,µ i,! i } n M: number of mixtures ( ) n F: feature dimension (20-60) n MF ( ~20k- 50k) 9/31

10 Eigenvoices & Eigenchannels GMM-UBM (MAP): sh = + Dz sh D: Full-rank diagonal (scaling factor) z: speaker component s speaker h utterance µ UBM supervector µ s speaker supervector µ sh target model supervector Eigenvoices: s = + Vy s V: speaker variability subspace (low-rank) y: corresponding weights for a given speaker, speaker factors Eigenchannels sh = s + Ux h U: session variability subspace (low-rank) x: corresponding weights for a given utterance/speaker, channel factors 10/31

11 Joint Factor Analysis: Eigenvoices + Eigenchannels + MAP Model = Speaker/language + Session [Kenny 04]. μ s s speaker h utterance µ UBM supervector µ s speaker supervector µ sh target model supervector V speaker variability subspace U session variability subspace x channel factors y speaker factors µ sh = µ + Vy s + Dz s + Ux sh 11/31

12 Outline 1. The Inter-session Variability Problem 2. From Eigenfaces to Joint Factor Analysis 3. Factor Analysis in Speaker and Language Recognition: I. Theory II. Where and How III. Efficiency 4. Total Variability 5. PLDA 6. A link with new machine learning paradigms 7. Results (NIST SRE10, SRE 12) 12/31

13 Factor Analysis: Graphical model! z n x n = observed variables (speaker supervectors). z n = latent variables (channel or speaker factors). µ L Hyperparameters (, L, ) = UBM supervector. L = U, V = UBM covariance x n N 13/31

14 Joint Factor Analysis: Point estimate of latent factors, x, y n A point essmate (mean of posterior) of x, y can be computed as in classic relevance MAP![z x] = " f! = (I + L T N" #1 $ #1 L) #1 L T $ #1 p(z) E[z x] p(z x) x 1 P(x z) ~ N (μ + Lz, Ψ) 0 Latent Variables Domain (D =1) Observations Domain (D = 2) x 2 14/31

15 Outline 1. The Inter-session Variability Problem 2. From Eigenfaces to Joint Factor Analysis 3. Factor Analysis in Speaker and Language Recognition: I. Theory II. Where and How III. Efficiency 4. Total Variability 5. PLDA 6. Results (NIST SRE10, SRE 12) 15/31

16 Factor Analysis: Where and How n Two classifiers (Acoustic systems) GMM SVM Maximum Margin Hyperplane Support Vectors Support Vectors Margin n Three different levels (domains) Feature domain Statistics domain Model domain (supervectors) 16/31

17 Outline 1. The Inter-session Variability Problem 2. From Eigenfaces to Joint Factor Analysis 3. Factor Analysis in Speaker and Language Recognition: I. Theory II. Where and How III. Efficiency 4. Total Variability 5. PLDA 6. Results (NIST SRE10, SRE 12) 17/31

18 Efficiency: ATVS system at NIST SRE2008 tel-tel Step System SVM JFA JFA- LS Development UBM training (2M feature vectors, gender dependent) 4h 4h 4h Training Variability Subspace U/V 1h 1h/1h 1h/1h Feature extrachon (per ~265s file) MFCC 2s 2s 2s Training (per ~265s file) GMM- train 8s 8s 8s FA point- essmate 0.1s 0.1s SVM- train 120s Total(train) 130.1s 10.1s 10.1s xrt train (CPU/speech) 0.50RT 0.04RT 0.04RT TesHng (per ~265s file) SV- train 8s FA point- essmate 0.1s 0.1s Scoring(frame by frame/ linear scoring) 3.2s 0.2s 1 x 10-4 s t- norm(100 models) 320s 20s 1x10-2 s Total(test) 331.2s 22.2s 2.02s xrt test (CPU/speech) 1.24RT 0.08RT 7.5x10-3 RT Training: 1min speech is processed in SVM: 30s JFA: 2.4s JFA-LS: 2.4s Testing: 1min speech is processed in SVM: 74.4s JFA: 4.8s JFA-LS: 0.45s 18/31

19 Outline 1. The Inter-session Variability Problem 2. From Eigenfaces to Joint Factor Analysis 3. Factor Analysis in Speaker and Language Recognition: I. Theory II. Where and How III. Efficiency 4. Total Variability 5. PLDA 6. Results (NIST SRE10, SRE 12) 19/31

20 Total Variability : m = m + Tw n Limited real data restrictions à U estimation might include speaker information. n Total Variability: T represents both session and target information m s = m + Tw sh n Disentangling phase in w domain (LDA, WCCN) 20/31

21 Outline 1. The Inter-session Variability Problem 2. From Eigenfaces to Joint Factor Analysis 3. Factor Analysis in Speaker and Language Recognition: I. Theory II. Where and How III. Efficiency 4. Total Variability 5. PLDA 6. Results (NIST SRE10, SRE 12) 21/31

22 PLDA: FA over i-vectors W ij i speaker j utterance w mean i-vector w ij target model i-vector F speaker variability subspace G session variability subspace k channel factors h speaker factors e noise term w ij = w + Fh i + Gk ij + e ij 22/31

23 PLDA: FA over i-vectors I speaker H utterance w i-vector F speaker variability subspace G session variability subspace k channel factors h speaker factors e noise term h k w H I Θ = {μ,f,g,σ} 23/31

24 Outline 1. The Inter-session Variability Problem 2. From Eigenfaces to Joint Factor Analysis 3. Factor Analysis in Speaker and Language Recognition: I. Theory II. Where and How III. Efficiency 4. Total Variability 5. PLDA 6. Results (NIST SRE10, SRE 12) 24/31

25 Experimental Results: SRE 10 n Organizer n n National Institute of Standards and Technology (NIST) Competitive participants MIT-LL, SRI, IBM Relevant data about task 2 Channel types involved (telephone, microphone) 2 Speech style involved (conversational, interview) Different vocal effort (high, neutral, low) ~150s train/test ~ 900 speakers ~ 800 models ~ 900 Test Files > Trials 25/31

26 Experimental Results: SRE 10, some samples n Telephone data n Microphone data Close mic Far mic n Vocal Effort Low Vocal Effort High Vocal Effort 26/31

27 Experimental Results: SRE 10 CondiHon EER_male EER_female EER_all C01_ext int vs. int, matched mic 0,54 0,96 0,84 C02_ext int vs. int, mismatched mic 0,47 1,27 1,11 C03_ext int vs. tel, mic vs. phn 1,88 3,54 2,71 C04_ext int vs. tel, mic vs. mic 1,37 3,69 2,52 C05_ext tel vs. tel, phn vs. phn 2,02 2,36 2,22 C06_ext tel vs. tel, normal vs. high vel 2,9 3,54 3,37 C07_ext mic vs. mic, normal vs. high vel 4,03 4,54 4,27 C08_ext tel vs. tel, normal vs. low vel 1,34 1,86 1,69 C09_ext mic vs. mic, normal vs. low vel 4,37 3,56 4,58 27/31

28 Experimental Results: SRE 12 n Drastic changes from past sre evaluations Multitraining (different number of files for model training) Test Variability Duration (20s to 160s) Noisy conditions n Large amount of test files under noisy conditions (10dbs, 0dbs SNR ) n Reverberation n An industrial task ~ 2K speakers ~ 1.8K models ~ 2.5K Test Files ~ 2M Trials (core); ~88M 28/31

29 Experimental Results: SRE 12, some samples n Noisy Files 10 dbs 0 dbs 29/31

30 Experimental Results: SRE 12, noise robustness 30/31

31 QUESTIONS 31/31

Joint Factor Analysis for Speaker Verification

Joint Factor Analysis for Speaker Verification Joint Factor Analysis for Speaker Verification Mengke HU ASPITRG Group, ECE Department Drexel University mengke.hu@gmail.com October 12, 2012 1/37 Outline 1 Speaker Verification Baseline System Session

More information

Front-End Factor Analysis For Speaker Verification

Front-End Factor Analysis For Speaker Verification IEEE TRANSACTIONS ON AUDIO, SPEECH AND LANGUAGE PROCESSING Front-End Factor Analysis For Speaker Verification Najim Dehak, Patrick Kenny, Réda Dehak, Pierre Dumouchel, and Pierre Ouellet, Abstract This

More information

i-vector and GMM-UBM Bie Fanhu CSLT, RIIT, THU

i-vector and GMM-UBM Bie Fanhu CSLT, RIIT, THU i-vector and GMM-UBM Bie Fanhu CSLT, RIIT, THU 2013-11-18 Framework 1. GMM-UBM Feature is extracted by frame. Number of features are unfixed. Gaussian Mixtures are used to fit all the features. The mixtures

More information

Support Vector Machines using GMM Supervectors for Speaker Verification

Support Vector Machines using GMM Supervectors for Speaker Verification 1 Support Vector Machines using GMM Supervectors for Speaker Verification W. M. Campbell, D. E. Sturim, D. A. Reynolds MIT Lincoln Laboratory 244 Wood Street Lexington, MA 02420 Corresponding author e-mail:

More information

Uncertainty Modeling without Subspace Methods for Text-Dependent Speaker Recognition

Uncertainty Modeling without Subspace Methods for Text-Dependent Speaker Recognition Uncertainty Modeling without Subspace Methods for Text-Dependent Speaker Recognition Patrick Kenny, Themos Stafylakis, Md. Jahangir Alam and Marcel Kockmann Odyssey Speaker and Language Recognition Workshop

More information

A Small Footprint i-vector Extractor

A Small Footprint i-vector Extractor A Small Footprint i-vector Extractor Patrick Kenny Odyssey Speaker and Language Recognition Workshop June 25, 2012 1 / 25 Patrick Kenny A Small Footprint i-vector Extractor Outline Introduction Review

More information

Speaker Verification Using Accumulative Vectors with Support Vector Machines

Speaker Verification Using Accumulative Vectors with Support Vector Machines Speaker Verification Using Accumulative Vectors with Support Vector Machines Manuel Aguado Martínez, Gabriel Hernández-Sierra, and José Ramón Calvo de Lara Advanced Technologies Application Center, Havana,

More information

TNO SRE-2008: Calibration over all trials and side-information

TNO SRE-2008: Calibration over all trials and side-information Image from Dr Seuss TNO SRE-2008: Calibration over all trials and side-information David van Leeuwen (TNO, ICSI) Howard Lei (ICSI), Nir Krause (PRS), Albert Strasheim (SUN) Niko Brümmer (SDV) Knowledge

More information

An Integration of Random Subspace Sampling and Fishervoice for Speaker Verification

An Integration of Random Subspace Sampling and Fishervoice for Speaker Verification Odyssey 2014: The Speaker and Language Recognition Workshop 16-19 June 2014, Joensuu, Finland An Integration of Random Subspace Sampling and Fishervoice for Speaker Verification Jinghua Zhong 1, Weiwu

More information

Low-dimensional speech representation based on Factor Analysis and its applications!

Low-dimensional speech representation based on Factor Analysis and its applications! Low-dimensional speech representation based on Factor Analysis and its applications! Najim Dehak and Stephen Shum! Spoken Language System Group! MIT Computer Science and Artificial Intelligence Laboratory!

More information

Speaker recognition by means of Deep Belief Networks

Speaker recognition by means of Deep Belief Networks Speaker recognition by means of Deep Belief Networks Vasileios Vasilakakis, Sandro Cumani, Pietro Laface, Politecnico di Torino, Italy {first.lastname}@polito.it 1. Abstract Most state of the art speaker

More information

Modified-prior PLDA and Score Calibration for Duration Mismatch Compensation in Speaker Recognition System

Modified-prior PLDA and Score Calibration for Duration Mismatch Compensation in Speaker Recognition System INERSPEECH 2015 Modified-prior PLDA and Score Calibration for Duration Mismatch Compensation in Speaker Recognition System QingYang Hong 1, Lin Li 1, Ming Li 2, Ling Huang 1, Lihong Wan 1, Jun Zhang 1

More information

Novel Quality Metric for Duration Variability Compensation in Speaker Verification using i-vectors

Novel Quality Metric for Duration Variability Compensation in Speaker Verification using i-vectors Published in Ninth International Conference on Advances in Pattern Recognition (ICAPR-2017), Bangalore, India Novel Quality Metric for Duration Variability Compensation in Speaker Verification using i-vectors

More information

EFFECTIVE ACOUSTIC MODELING FOR ROBUST SPEAKER RECOGNITION. Taufiq Hasan Al Banna

EFFECTIVE ACOUSTIC MODELING FOR ROBUST SPEAKER RECOGNITION. Taufiq Hasan Al Banna EFFECTIVE ACOUSTIC MODELING FOR ROBUST SPEAKER RECOGNITION by Taufiq Hasan Al Banna APPROVED BY SUPERVISORY COMMITTEE: Dr. John H. L. Hansen, Chair Dr. Carlos Busso Dr. Hlaing Minn Dr. P. K. Rajasekaran

More information

INTERSPEECH 2016 Tutorial: Machine Learning for Speaker Recognition

INTERSPEECH 2016 Tutorial: Machine Learning for Speaker Recognition INTERSPEECH 2016 Tutorial: Machine Learning for Speaker Recognition Man-Wai Mak and Jen-Tzung Chien The Hong Kong Polytechnic University, Hong Kong National Chiao Tung University, Taiwan September 8, 2016

More information

Multiclass Discriminative Training of i-vector Language Recognition

Multiclass Discriminative Training of i-vector Language Recognition Odyssey 214: The Speaker and Language Recognition Workshop 16-19 June 214, Joensuu, Finland Multiclass Discriminative Training of i-vector Language Recognition Alan McCree Human Language Technology Center

More information

Usually the estimation of the partition function is intractable and it becomes exponentially hard when the complexity of the model increases. However,

Usually the estimation of the partition function is intractable and it becomes exponentially hard when the complexity of the model increases. However, Odyssey 2012 The Speaker and Language Recognition Workshop 25-28 June 2012, Singapore First attempt of Boltzmann Machines for Speaker Verification Mohammed Senoussaoui 1,2, Najim Dehak 3, Patrick Kenny

More information

IBM Research Report. Training Universal Background Models for Speaker Recognition

IBM Research Report. Training Universal Background Models for Speaker Recognition RC24953 (W1003-002) March 1, 2010 Other IBM Research Report Training Universal Bacground Models for Speaer Recognition Mohamed Kamal Omar, Jason Pelecanos IBM Research Division Thomas J. Watson Research

More information

SCORE CALIBRATING FOR SPEAKER RECOGNITION BASED ON SUPPORT VECTOR MACHINES AND GAUSSIAN MIXTURE MODELS

SCORE CALIBRATING FOR SPEAKER RECOGNITION BASED ON SUPPORT VECTOR MACHINES AND GAUSSIAN MIXTURE MODELS SCORE CALIBRATING FOR SPEAKER RECOGNITION BASED ON SUPPORT VECTOR MACHINES AND GAUSSIAN MIXTURE MODELS Marcel Katz, Martin Schafföner, Sven E. Krüger, Andreas Wendemuth IESK-Cognitive Systems University

More information

Independent Component Analysis and Unsupervised Learning. Jen-Tzung Chien

Independent Component Analysis and Unsupervised Learning. Jen-Tzung Chien Independent Component Analysis and Unsupervised Learning Jen-Tzung Chien TABLE OF CONTENTS 1. Independent Component Analysis 2. Case Study I: Speech Recognition Independent voices Nonparametric likelihood

More information

Independent Component Analysis and Unsupervised Learning

Independent Component Analysis and Unsupervised Learning Independent Component Analysis and Unsupervised Learning Jen-Tzung Chien National Cheng Kung University TABLE OF CONTENTS 1. Independent Component Analysis 2. Case Study I: Speech Recognition Independent

More information

Unifying Probabilistic Linear Discriminant Analysis Variants in Biometric Authentication

Unifying Probabilistic Linear Discriminant Analysis Variants in Biometric Authentication Unifying Probabilistic Linear Discriminant Analysis Variants in Biometric Authentication Aleksandr Sizov 1, Kong Aik Lee, Tomi Kinnunen 1 1 School of Computing, University of Eastern Finland, Finland Institute

More information

University of Birmingham Research Archive

University of Birmingham Research Archive University of Birmingham Research Archive e-theses repository This unpublished thesis/dissertation is copyright of the author and/or third parties. The intellectual property rights of the author or third

More information

Using Deep Belief Networks for Vector-Based Speaker Recognition

Using Deep Belief Networks for Vector-Based Speaker Recognition INTERSPEECH 2014 Using Deep Belief Networks for Vector-Based Speaker Recognition W. M. Campbell MIT Lincoln Laboratory, Lexington, MA, USA wcampbell@ll.mit.edu Abstract Deep belief networks (DBNs) have

More information

Heeyoul (Henry) Choi. Dept. of Computer Science Texas A&M University

Heeyoul (Henry) Choi. Dept. of Computer Science Texas A&M University Heeyoul (Henry) Choi Dept. of Computer Science Texas A&M University hchoi@cs.tamu.edu Introduction Speaker Adaptation Eigenvoice Comparison with others MAP, MLLR, EMAP, RMP, CAT, RSW Experiments Future

More information

Unsupervised Methods for Speaker Diarization. Stephen Shum

Unsupervised Methods for Speaker Diarization. Stephen Shum Unsupervised Methods for Speaker Diarization by Stephen Shum B.S., University of California, Berkeley (2009) Submitted to the Department of Electrical Engineering and Computer Science in partial fulfillment

More information

Support Vector Machines and Speaker Verification

Support Vector Machines and Speaker Verification 1 Support Vector Machines and Speaker Verification David Cinciruk March 6, 2013 2 Table of Contents Review of Speaker Verification Introduction to Support Vector Machines Derivation of SVM Equations Soft

More information

Around the Speaker De-Identification (Speaker diarization for de-identification ++) Itshak Lapidot Moez Ajili Jean-Francois Bonastre

Around the Speaker De-Identification (Speaker diarization for de-identification ++) Itshak Lapidot Moez Ajili Jean-Francois Bonastre Around the Speaker De-Identification (Speaker diarization for de-identification ++) Itshak Lapidot Moez Ajili Jean-Francois Bonastre The 2 Parts HDM based diarization System The homogeneity measure 2 Outline

More information

Minimax i-vector extractor for short duration speaker verification

Minimax i-vector extractor for short duration speaker verification Minimax i-vector extractor for short duration speaker verification Ville Hautamäki 1,2, You-Chi Cheng 2, Padmanabhan Rajan 1, Chin-Hui Lee 2 1 School of Computing, University of Eastern Finl, Finl 2 ECE,

More information

Fast speaker diarization based on binary keys. Xavier Anguera and Jean François Bonastre

Fast speaker diarization based on binary keys. Xavier Anguera and Jean François Bonastre Fast speaker diarization based on binary keys Xavier Anguera and Jean François Bonastre Outline Introduction Speaker diarization Binary speaker modeling Binary speaker diarization system Experiments Conclusions

More information

Harmonic Structure Transform for Speaker Recognition

Harmonic Structure Transform for Speaker Recognition Harmonic Structure Transform for Speaker Recognition Kornel Laskowski & Qin Jin Carnegie Mellon University, Pittsburgh PA, USA KTH Speech Music & Hearing, Stockholm, Sweden 29 August, 2011 Laskowski &

More information

Robust Speaker Identification

Robust Speaker Identification Robust Speaker Identification by Smarajit Bose Interdisciplinary Statistical Research Unit Indian Statistical Institute, Kolkata Joint work with Amita Pal and Ayanendranath Basu Overview } } } } } } }

More information

Improving the Effectiveness of Speaker Verification Domain Adaptation With Inadequate In-Domain Data

Improving the Effectiveness of Speaker Verification Domain Adaptation With Inadequate In-Domain Data Distribution A: Public Release Improving the Effectiveness of Speaker Verification Domain Adaptation With Inadequate In-Domain Data Bengt J. Borgström Elliot Singer Douglas Reynolds and Omid Sadjadi 2

More information

Noise Compensation for Subspace Gaussian Mixture Models

Noise Compensation for Subspace Gaussian Mixture Models Noise ompensation for ubspace Gaussian Mixture Models Liang Lu University of Edinburgh Joint work with KK hin, A. Ghoshal and. enals Liang Lu, Interspeech, eptember, 2012 Outline Motivation ubspace GMM

More information

Monaural speech separation using source-adapted models

Monaural speech separation using source-adapted models Monaural speech separation using source-adapted models Ron Weiss, Dan Ellis {ronw,dpwe}@ee.columbia.edu LabROSA Department of Electrical Enginering Columbia University 007 IEEE Workshop on Applications

More information

Automatic Speech Recognition (CS753)

Automatic Speech Recognition (CS753) Automatic Speech Recognition (CS753) Lecture 21: Speaker Adaptation Instructor: Preethi Jyothi Oct 23, 2017 Speaker variations Major cause of variability in speech is the differences between speakers Speaking

More information

Introduction to Support Vector Machines

Introduction to Support Vector Machines Introduction to Support Vector Machines Hsuan-Tien Lin Learning Systems Group, California Institute of Technology Talk in NTU EE/CS Speech Lab, November 16, 2005 H.-T. Lin (Learning Systems Group) Introduction

More information

Mixtures of Gaussians with Sparse Regression Matrices. Constantinos Boulis, Jeffrey Bilmes

Mixtures of Gaussians with Sparse Regression Matrices. Constantinos Boulis, Jeffrey Bilmes Mixtures of Gaussians with Sparse Regression Matrices Constantinos Boulis, Jeffrey Bilmes {boulis,bilmes}@ee.washington.edu Dept of EE, University of Washington Seattle WA, 98195-2500 UW Electrical Engineering

More information

Spectral and Textural Feature-Based System for Automatic Detection of Fricatives and Affricates

Spectral and Textural Feature-Based System for Automatic Detection of Fricatives and Affricates Spectral and Textural Feature-Based System for Automatic Detection of Fricatives and Affricates Dima Ruinskiy Niv Dadush Yizhar Lavner Department of Computer Science, Tel-Hai College, Israel Outline Phoneme

More information

A Generative Model Based Kernel for SVM Classification in Multimedia Applications

A Generative Model Based Kernel for SVM Classification in Multimedia Applications Appears in Neural Information Processing Systems, Vancouver, Canada, 2003. A Generative Model Based Kernel for SVM Classification in Multimedia Applications Pedro J. Moreno Purdy P. Ho Hewlett-Packard

More information

ISCA Archive

ISCA Archive ISCA Archive http://www.isca-speech.org/archive ODYSSEY04 - The Speaker and Language Recognition Workshop Toledo, Spain May 3 - June 3, 2004 Analysis of Multitarget Detection for Speaker and Language Recognition*

More information

Lecture 7: Con3nuous Latent Variable Models

Lecture 7: Con3nuous Latent Variable Models CSC2515 Fall 2015 Introduc3on to Machine Learning Lecture 7: Con3nuous Latent Variable Models All lecture slides will be available as.pdf on the course website: http://www.cs.toronto.edu/~urtasun/courses/csc2515/

More information

Eigenvoice Speaker Adaptation via Composite Kernel PCA

Eigenvoice Speaker Adaptation via Composite Kernel PCA Eigenvoice Speaker Adaptation via Composite Kernel PCA James T. Kwok, Brian Mak and Simon Ho Department of Computer Science Hong Kong University of Science and Technology Clear Water Bay, Hong Kong [jamesk,mak,csho]@cs.ust.hk

More information

Studies on Model Distance Normalization Approach in Text-independent Speaker Verification

Studies on Model Distance Normalization Approach in Text-independent Speaker Verification Vol. 35, No. 5 ACTA AUTOMATICA SINICA May, 009 Studies on Model Distance Normalization Approach in Text-independent Speaker Verification DONG Yuan LU Liang ZHAO Xian-Yu ZHAO Jian Abstract Model distance

More information

PCA & ICA. CE-717: Machine Learning Sharif University of Technology Spring Soleymani

PCA & ICA. CE-717: Machine Learning Sharif University of Technology Spring Soleymani PCA & ICA CE-717: Machine Learning Sharif University of Technology Spring 2015 Soleymani Dimensionality Reduction: Feature Selection vs. Feature Extraction Feature selection Select a subset of a given

More information

Spoken Language Understanding in a Latent Topic-based Subspace

Spoken Language Understanding in a Latent Topic-based Subspace INTERSPEECH 2016 September 8 12, 2016, San Francisco, USA Spoken Language Understanding in a Latent Topic-based Subspace Mohamed Morchid 1, Mohamed Bouaziz 1,3, Waad Ben Kheder 1, Killian Janod 1,2, Pierre-Michel

More information

The effect of speaking rate and vowel context on the perception of consonants. in babble noise

The effect of speaking rate and vowel context on the perception of consonants. in babble noise The effect of speaking rate and vowel context on the perception of consonants in babble noise Anirudh Raju Department of Electrical Engineering, University of California, Los Angeles, California, USA anirudh90@ucla.edu

More information

Approximate Bayesian Inference for Robust Speech Processing. A Thesis. Submitted to the Faculty. Drexel University. Ciira wa Maina

Approximate Bayesian Inference for Robust Speech Processing. A Thesis. Submitted to the Faculty. Drexel University. Ciira wa Maina Approximate Bayesian Inference for Robust Speech Processing A Thesis Submitted to the Faculty of Drexel University by Ciira wa Maina in partial fulfillment of the requirements for the degree of Doctor

More information

Estimation of Relative Operating Characteristics of Text Independent Speaker Verification

Estimation of Relative Operating Characteristics of Text Independent Speaker Verification International Journal of Engineering Science Invention Volume 1 Issue 1 December. 2012 PP.18-23 Estimation of Relative Operating Characteristics of Text Independent Speaker Verification Palivela Hema 1,

More information

Bayesian Analysis of Speaker Diarization with Eigenvoice Priors

Bayesian Analysis of Speaker Diarization with Eigenvoice Priors Bayesian Analysis of Speaker Diarization with Eigenvoice Priors Patrick Kenny Centre de recherche informatique de Montréal Patrick.Kenny@crim.ca A year in the lab can save you a day in the library. Panu

More information

Gain Compensation for Fast I-Vector Extraction over Short Duration

Gain Compensation for Fast I-Vector Extraction over Short Duration INTERSPEECH 27 August 2 24, 27, Stockholm, Sweden Gain Compensation for Fast I-Vector Extraction over Short Duration Kong Aik Lee and Haizhou Li 2 Institute for Infocomm Research I 2 R), A STAR, Singapore

More information

Spatial Diffuseness Features for DNN-Based Speech Recognition in Noisy and Reverberant Environments

Spatial Diffuseness Features for DNN-Based Speech Recognition in Noisy and Reverberant Environments Spatial Diffuseness Features for DNN-Based Speech Recognition in Noisy and Reverberant Environments Andreas Schwarz, Christian Huemmer, Roland Maas, Walter Kellermann Lehrstuhl für Multimediakommunikation

More information

Gaussian Mixture Model Uncertainty Learning (GMMUL) Version 1.0 User Guide

Gaussian Mixture Model Uncertainty Learning (GMMUL) Version 1.0 User Guide Gaussian Mixture Model Uncertainty Learning (GMMUL) Version 1. User Guide Alexey Ozerov 1, Mathieu Lagrange and Emmanuel Vincent 1 1 INRIA, Centre de Rennes - Bretagne Atlantique Campus de Beaulieu, 3

More information

Environmental Sound Classification in Realistic Situations

Environmental Sound Classification in Realistic Situations Environmental Sound Classification in Realistic Situations K. Haddad, W. Song Brüel & Kjær Sound and Vibration Measurement A/S, Skodsborgvej 307, 2850 Nærum, Denmark. X. Valero La Salle, Universistat Ramon

More information

Maximum Likelihood and Maximum A Posteriori Adaptation for Distributed Speaker Recognition Systems

Maximum Likelihood and Maximum A Posteriori Adaptation for Distributed Speaker Recognition Systems Maximum Likelihood and Maximum A Posteriori Adaptation for Distributed Speaker Recognition Systems Chin-Hung Sit 1, Man-Wai Mak 1, and Sun-Yuan Kung 2 1 Center for Multimedia Signal Processing Dept. of

More information

Advanced Introduction to Machine Learning

Advanced Introduction to Machine Learning 10-715 Advanced Introduction to Machine Learning Homework 3 Due Nov 12, 10.30 am Rules 1. Homework is due on the due date at 10.30 am. Please hand over your homework at the beginning of class. Please see

More information

Domain-invariant I-vector Feature Extraction for PLDA Speaker Verification

Domain-invariant I-vector Feature Extraction for PLDA Speaker Verification Odyssey 2018 The Speaker and Language Recognition Workshop 26-29 June 2018, Les Sables d Olonne, France Domain-invariant I-vector Feature Extraction for PLDA Speaker Verification Md Hafizur Rahman 1, Ivan

More information

Automatic Regularization of Cross-entropy Cost for Speaker Recognition Fusion

Automatic Regularization of Cross-entropy Cost for Speaker Recognition Fusion INTERSPEECH 203 Automatic Regularization of Cross-entropy Cost for Speaker Recognition Fusion Ville Hautamäki, Kong Aik Lee 2, David van Leeuwen 3, Rahim Saeidi 3, Anthony Larcher 2, Tomi Kinnunen, Taufiq

More information

A latent variable modelling approach to the acoustic-to-articulatory mapping problem

A latent variable modelling approach to the acoustic-to-articulatory mapping problem A latent variable modelling approach to the acoustic-to-articulatory mapping problem Miguel Á. Carreira-Perpiñán and Steve Renals Dept. of Computer Science, University of Sheffield {miguel,sjr}@dcs.shef.ac.uk

More information

Computer Vision Group Prof. Daniel Cremers. 6. Mixture Models and Expectation-Maximization

Computer Vision Group Prof. Daniel Cremers. 6. Mixture Models and Expectation-Maximization Prof. Daniel Cremers 6. Mixture Models and Expectation-Maximization Motivation Often the introduction of latent (unobserved) random variables into a model can help to express complex (marginal) distributions

More information

Eigenface-based facial recognition

Eigenface-based facial recognition Eigenface-based facial recognition Dimitri PISSARENKO December 1, 2002 1 General This document is based upon Turk and Pentland (1991b), Turk and Pentland (1991a) and Smith (2002). 2 How does it work? The

More information

Full-covariance model compensation for

Full-covariance model compensation for compensation transms Presentation Toshiba, 12 Mar 2008 Outline compensation transms compensation transms Outline compensation transms compensation transms Noise model x clean speech; n additive ; h convolutional

More information

Fuzzy Support Vector Machines for Automatic Infant Cry Recognition

Fuzzy Support Vector Machines for Automatic Infant Cry Recognition Fuzzy Support Vector Machines for Automatic Infant Cry Recognition Sandra E. Barajas-Montiel and Carlos A. Reyes-García Instituto Nacional de Astrofisica Optica y Electronica, Luis Enrique Erro #1, Tonantzintla,

More information

Machine Learning. B. Unsupervised Learning B.2 Dimensionality Reduction. Lars Schmidt-Thieme, Nicolas Schilling

Machine Learning. B. Unsupervised Learning B.2 Dimensionality Reduction. Lars Schmidt-Thieme, Nicolas Schilling Machine Learning B. Unsupervised Learning B.2 Dimensionality Reduction Lars Schmidt-Thieme, Nicolas Schilling Information Systems and Machine Learning Lab (ISMLL) Institute for Computer Science University

More information

Augmented Statistical Models for Speech Recognition

Augmented Statistical Models for Speech Recognition Augmented Statistical Models for Speech Recognition Mark Gales & Martin Layton 31 August 2005 Trajectory Models For Speech Processing Workshop Overview Dependency Modelling in Speech Recognition: latent

More information

Engineering Part IIB: Module 4F11 Speech and Language Processing Lectures 4/5 : Speech Recognition Basics

Engineering Part IIB: Module 4F11 Speech and Language Processing Lectures 4/5 : Speech Recognition Basics Engineering Part IIB: Module 4F11 Speech and Language Processing Lectures 4/5 : Speech Recognition Basics Phil Woodland: pcw@eng.cam.ac.uk Lent 2013 Engineering Part IIB: Module 4F11 What is Speech Recognition?

More information

PCA and LDA. Man-Wai MAK

PCA and LDA. Man-Wai MAK PCA and LDA Man-Wai MAK Dept. of Electronic and Information Engineering, The Hong Kong Polytechnic University enmwmak@polyu.edu.hk http://www.eie.polyu.edu.hk/ mwmak References: S.J.D. Prince,Computer

More information

A Coupled Helmholtz Machine for PCA

A Coupled Helmholtz Machine for PCA A Coupled Helmholtz Machine for PCA Seungjin Choi Department of Computer Science Pohang University of Science and Technology San 3 Hyoja-dong, Nam-gu Pohang 79-784, Korea seungjin@postech.ac.kr August

More information

Segmental Recurrent Neural Networks for End-to-end Speech Recognition

Segmental Recurrent Neural Networks for End-to-end Speech Recognition Segmental Recurrent Neural Networks for End-to-end Speech Recognition Liang Lu, Lingpeng Kong, Chris Dyer, Noah Smith and Steve Renals TTI-Chicago, UoE, CMU and UW 9 September 2016 Background A new wave

More information

Factor Analysis based Semantic Variability Compensation for Automatic Conversation Representation

Factor Analysis based Semantic Variability Compensation for Automatic Conversation Representation Factor Analysis based Semantic Variability Compensation for Automatic Conversation Representation Mohamed Bouallegue, Mohamed Morchid, Richard Dufour, Driss Matrouf, Georges Linarès and Renato De Mori,

More information

Sparse Models for Speech Recognition

Sparse Models for Speech Recognition Sparse Models for Speech Recognition Weibin Zhang and Pascale Fung Human Language Technology Center Hong Kong University of Science and Technology Outline Introduction to speech recognition Motivations

More information

Machine Recognition of Sounds in Mixtures

Machine Recognition of Sounds in Mixtures Machine Recognition of Sounds in Mixtures Outline 1 2 3 4 Computational Auditory Scene Analysis Speech Recognition as Source Formation Sound Fragment Decoding Results & Conclusions Dan Ellis

More information

A Variance Modeling Framework Based on Variational Autoencoders for Speech Enhancement

A Variance Modeling Framework Based on Variational Autoencoders for Speech Enhancement A Variance Modeling Framework Based on Variational Autoencoders for Speech Enhancement Simon Leglaive 1 Laurent Girin 1,2 Radu Horaud 1 1: Inria Grenoble Rhône-Alpes 2: Univ. Grenoble Alpes, Grenoble INP,

More information

Clustering K-means. Clustering images. Machine Learning CSE546 Carlos Guestrin University of Washington. November 4, 2014.

Clustering K-means. Clustering images. Machine Learning CSE546 Carlos Guestrin University of Washington. November 4, 2014. Clustering K-means Machine Learning CSE546 Carlos Guestrin University of Washington November 4, 2014 1 Clustering images Set of Images [Goldberger et al.] 2 1 K-means Randomly initialize k centers µ (0)

More information

Model-based unsupervised segmentation of birdcalls from field recordings

Model-based unsupervised segmentation of birdcalls from field recordings Model-based unsupervised segmentation of birdcalls from field recordings Anshul Thakur School of Computing and Electrical Engineering Indian Institute of Technology Mandi Himachal Pradesh, India Email:

More information

CS281 Section 4: Factor Analysis and PCA

CS281 Section 4: Factor Analysis and PCA CS81 Section 4: Factor Analysis and PCA Scott Linderman At this point we have seen a variety of machine learning models, with a particular emphasis on models for supervised learning. In particular, we

More information

Reformulating the HMM as a trajectory model by imposing explicit relationship between static and dynamic features

Reformulating the HMM as a trajectory model by imposing explicit relationship between static and dynamic features Reformulating the HMM as a trajectory model by imposing explicit relationship between static and dynamic features Heiga ZEN (Byung Ha CHUN) Nagoya Inst. of Tech., Japan Overview. Research backgrounds 2.

More information

Discriminative Direction for Kernel Classifiers

Discriminative Direction for Kernel Classifiers Discriminative Direction for Kernel Classifiers Polina Golland Artificial Intelligence Lab Massachusetts Institute of Technology Cambridge, MA 02139 polina@ai.mit.edu Abstract In many scientific and engineering

More information

L5 Support Vector Classification

L5 Support Vector Classification L5 Support Vector Classification Support Vector Machine Problem definition Geometrical picture Optimization problem Optimization Problem Hard margin Convexity Dual problem Soft margin problem Alexander

More information

Pattern Classification

Pattern Classification Pattern Classification Introduction Parametric classifiers Semi-parametric classifiers Dimensionality reduction Significance testing 6345 Automatic Speech Recognition Semi-Parametric Classifiers 1 Semi-Parametric

More information

Speaker Representation and Verification Part II. by Vasileios Vasilakakis

Speaker Representation and Verification Part II. by Vasileios Vasilakakis Speaker Representation and Verification Part II by Vasileios Vasilakakis Outline -Approaches of Neural Networks in Speaker/Speech Recognition -Feed-Forward Neural Networks -Training with Back-propagation

More information

Kernel Methods for Text-Independent Speaker Verification

Kernel Methods for Text-Independent Speaker Verification Kernel Methods for Text-Independent Speaker Verification Chris Longworth Cambridge University Engineering Department and Christ s College February 25, 2010 Dissertation submitted to the University of Cambridge

More information

Analysis of mutual duration and noise effects in speaker recognition: benefits of condition-matched cohort selection in score normalization

Analysis of mutual duration and noise effects in speaker recognition: benefits of condition-matched cohort selection in score normalization Analysis of mutual duration and noise effects in speaker recognition: benefits of condition-matched cohort selection in score normalization Andreas Nautsch, Rahim Saeidi, Christian Rathgeb, and Christoph

More information

Exemplar-based voice conversion using non-negative spectrogram deconvolution

Exemplar-based voice conversion using non-negative spectrogram deconvolution Exemplar-based voice conversion using non-negative spectrogram deconvolution Zhizheng Wu 1, Tuomas Virtanen 2, Tomi Kinnunen 3, Eng Siong Chng 1, Haizhou Li 1,4 1 Nanyang Technological University, Singapore

More information

HMM part 1. Dr Philip Jackson

HMM part 1. Dr Philip Jackson Centre for Vision Speech & Signal Processing University of Surrey, Guildford GU2 7XH. HMM part 1 Dr Philip Jackson Probability fundamentals Markov models State topology diagrams Hidden Markov models -

More information

SPEECH recognition systems based on hidden Markov

SPEECH recognition systems based on hidden Markov IEEE SIGNAL PROCESSING LETTERS, VOL. X, NO. X, 2014 1 Probabilistic Linear Discriminant Analysis for Acoustic Modelling Liang Lu, Member, IEEE and Steve Renals, Fellow, IEEE Abstract In this letter, we

More information

Variational Autoencoders

Variational Autoencoders Variational Autoencoders Recap: Story so far A classification MLP actually comprises two components A feature extraction network that converts the inputs into linearly separable features Or nearly linearly

More information

Unsupervised Learning

Unsupervised Learning 2018 EE448, Big Data Mining, Lecture 7 Unsupervised Learning Weinan Zhang Shanghai Jiao Tong University http://wnzhang.net http://wnzhang.net/teaching/ee448/index.html ML Problem Setting First build and

More information

Modeling Prosody for Speaker Recognition: Why Estimating Pitch May Be a Red Herring

Modeling Prosody for Speaker Recognition: Why Estimating Pitch May Be a Red Herring Modeling Prosody for Speaker Recognition: Why Estimating Pitch May Be a Red Herring Kornel Laskowski & Qin Jin Carnegie Mellon University Pittsburgh PA, USA 28 June, 2010 Laskowski & Jin ODYSSEY 2010,

More information

Noise Classification based on PCA. Nattanun Thatphithakkul, Boontee Kruatrachue, Chai Wutiwiwatchai, Vataya Boonpiam

Noise Classification based on PCA. Nattanun Thatphithakkul, Boontee Kruatrachue, Chai Wutiwiwatchai, Vataya Boonpiam Noise Classification based on PCA Nattanun Thatphithakkul, Boontee Kruatrachue, Chai Wutiwiwatchai, Vataya Boonpiam 1 Outline Introduction Principle component analysis (PCA) Classification using PCA Experiment

More information

PCA and LDA. Man-Wai MAK

PCA and LDA. Man-Wai MAK PCA and LDA Man-Wai MAK Dept. of Electronic and Information Engineering, The Hong Kong Polytechnic University enmwmak@polyu.edu.hk http://www.eie.polyu.edu.hk/ mwmak References: S.J.D. Prince,Computer

More information

Robust Sound Event Detection in Continuous Audio Environments

Robust Sound Event Detection in Continuous Audio Environments Robust Sound Event Detection in Continuous Audio Environments Haomin Zhang 1, Ian McLoughlin 2,1, Yan Song 1 1 National Engineering Laboratory of Speech and Language Information Processing The University

More information

Detection-Based Speech Recognition with Sparse Point Process Models

Detection-Based Speech Recognition with Sparse Point Process Models Detection-Based Speech Recognition with Sparse Point Process Models Aren Jansen Partha Niyogi Human Language Technology Center of Excellence Departments of Computer Science and Statistics ICASSP 2010 Dallas,

More information

Single Channel Signal Separation Using MAP-based Subspace Decomposition

Single Channel Signal Separation Using MAP-based Subspace Decomposition Single Channel Signal Separation Using MAP-based Subspace Decomposition Gil-Jin Jang, Te-Won Lee, and Yung-Hwan Oh 1 Spoken Language Laboratory, Department of Computer Science, KAIST 373-1 Gusong-dong,

More information

Note Set 5: Hidden Markov Models

Note Set 5: Hidden Markov Models Note Set 5: Hidden Markov Models Probabilistic Learning: Theory and Algorithms, CS 274A, Winter 2016 1 Hidden Markov Models (HMMs) 1.1 Introduction Consider observed data vectors x t that are d-dimensional

More information

SUBMITTED TO IEEE TRANSACTIONS ON SIGNAL PROCESSING 1. Correlation and Class Based Block Formation for Improved Structured Dictionary Learning

SUBMITTED TO IEEE TRANSACTIONS ON SIGNAL PROCESSING 1. Correlation and Class Based Block Formation for Improved Structured Dictionary Learning SUBMITTED TO IEEE TRANSACTIONS ON SIGNAL PROCESSING 1 Correlation and Class Based Block Formation for Improved Structured Dictionary Learning Nagendra Kumar and Rohit Sinha, Member, IEEE arxiv:178.1448v2

More information

Factor Analysis (10/2/13)

Factor Analysis (10/2/13) STA561: Probabilistic machine learning Factor Analysis (10/2/13) Lecturer: Barbara Engelhardt Scribes: Li Zhu, Fan Li, Ni Guan Factor Analysis Factor analysis is related to the mixture models we have studied.

More information

Multi-task Learning with Gaussian Processes, with Applications to Robot Inverse Dynamics

Multi-task Learning with Gaussian Processes, with Applications to Robot Inverse Dynamics 1 / 38 Multi-task Learning with Gaussian Processes, with Applications to Robot Inverse Dynamics Chris Williams with Kian Ming A. Chai, Stefan Klanke, Sethu Vijayakumar December 2009 Motivation 2 / 38 Examples

More information

STA 414/2104: Machine Learning

STA 414/2104: Machine Learning STA 414/2104: Machine Learning Russ Salakhutdinov Department of Computer Science! Department of Statistics! rsalakhu@cs.toronto.edu! http://www.cs.toronto.edu/~rsalakhu/ Lecture 8 Continuous Latent Variable

More information

Statistical Methods for SVM

Statistical Methods for SVM Statistical Methods for SVM Support Vector Machines Here we approach the two-class classification problem in a direct way: We try and find a plane that separates the classes in feature space. If we cannot,

More information