home committees paper program registration singapore participants plenary contactus

Odyssey 2012 Program

Monday 25 June 2012
8:30 - 9:30 Registration
9:30 - 10:00 Opening
10:00 - 11:00 Plenary Session 1: The Role of Proper Scoring Rules in Training and Evaluating Probabilistic Speaker and Language Recognizers
11:00 - 11:30 Coffee break
11:30 - 13:10 Session 1: Speaker Recognition – Compact Representation
13:10 - 14:00 Lunch
14:00 - 16:05 Session 2: Speaker Recognition – Generative modeling
16:05 - 16:30 Coffee break
16:30 - 17:45 Session 3: Forensic Speaker Recognition
17:45 - 22:00 Welcome Reception @ Tiger Brewery
Tuesday 26 June 2012
9:00 - 10:00 Plenary Session 2: Being Deep and Being Dynamic – New-Generation Models and Methodology for Advancing Speech Technology
10:00 - 10:30 Coffee break
10:30 - 12:35 Session 4: Neural Network for Speaker Recognition
12:35 - 13:30 Lunch
13:30 - 15:35 Session 5: Speaker Diarization
15:35 - 16:00 Coffee break
16:00 - 18:05 Session 6: Speaker Recognition – Channel Robustness
Wednesday 27 June 2012
9:00 - 10:00 Plenary Session 3: The NIST Speaker Recognition Evaluations
10:00 - 10:30 Coffee break
10:30 - 12:35 Session 7: Language Recognition Evaluation
12:35 - 13:30 Lunch
13:30 - 15:35 Session 8: Features for Speaker Recognition
15:35 - 22:00 Social Event + Banquet @ Megu
Thursday 28 June 2012
9:15 - 10:00 Coffee
10:00 - 12:05 Session 9: Speaker Recognition Evaluation
12:05 - 13:00 Lunch
13:00 - 15:05 Session 10: Speaker Recognition – Application
15:05 - 15:30 Coffee break
15:30 - 17:35 Session 11: Language Recognition – Feature, Classifier and Fusion

Plenary Session 1
Monday 25 June 2012, Session chair: Douglas Reynolds

10:00 - 11:00 The Role of Proper Scoring Rules in Training and Evaluating Probabilistic Speaker and Language Recognizers
Niko Brümmer

Plenary Session 2
Tuesday 26 June 2012, Session chair: Jean-Francois Bonastre

9:00 - 10:00 Being Deep and Being Dynamic – New-Generation Models and Methodology for Advancing Speech Technology
Li Deng

Plenary Session 3
Wednesday 27 June 2012, Session chair: Eliathamby Ambikairajah

9:00 - 10:00 The NIST Speaker Recognition Evaluations
Alvin Martin

Session 1: Speaker Recognition – Compact Representation
Monday 25 June 2012, Session chair: Lukáš Burget

11:30 - 11:55 A Small Footprint i-Vector Extractor
Patrick Kenny
11:55 - 12:20 Memory and Computation Effective Approaches for i–Vector Extraction
Sandro Cumani, Pietro Laface and Vasileios Vasilakakis
12:20 - 12:45 A Hybrid Factor Analysis and Probabilistic PCA-based system for Dictionary Learning and Encoding for Robust Speaker Recognition
Srikanth Madikeri
12:45 - 13:10 On Exploring the Similarity and Fusion of i-Vector and Sparse Representation based Speaker Verification Systems
Haris B C and Rohit Sinha

Session 2: Speaker Recognition – Generative modeling
Monday 25 June 2012, Session chair: Patrick Kenny

14:00 - 14:25 PLDA based Speaker Recognition on Short Utterances
Ahilan Kanagasundaram, Robbie Vogt, David Dean and Sridha Sridharan
14:25 - 14:50 PLDA based Speaker Verification with Weighted LDA Techniques
Ahilan Kanagasundaram, David Dean, Sridha Sridharan and Robbie Vogt
14:50 - 15:15 Dataset Shift in PLDA based Speaker Verification
Carlos Vaquero
15:15 - 15:40 Bayesian Adaptation of PLDA Based Speaker Recognition to Domains with Scarce Development Data
Jesus Villalba and Eduardo Lleida
15:40 - 16:05 Source Normalization for Language-Independent Speaker Recognition using i-Vectors
Mitchell McLaren, Miranti Indar Mandasari and David A. van Leeuwen

Session 3: Forensic Speaker Recognition
Monday 25 June 2012, Session chair: David van Leeuwen

16:30 - 16:55 Database Selection for Forensic Voice Comparison
Geoffrey Stewart Morrison, Felipe Ochoa, Tharmarajah Thiruvaran
16:55 - 17:20 Voice Source Features for Forensic Voice Comparison - an Evaluation of the GLOTTEX Software Package
Ewald Enzinger, Cuiling Zhang and Geoffrey Stewart Morrison
17:20 - 17:45 Comparison of Speaker Recognition Systems on a Real Forensic Benchmark
Yosef Solewicz, Timo Becker, Jardine Gaelle and Stefan Gfroerer

Session 4: Neural Network for Speaker Recognition
Tuesday 26 June 2012, Session chair: Hynek Hermansky

10:30 - 10:55 Factor Analysis of Mixture of Auto-Associative Neural Networks for Speaker Verification
Sivaram Garimella and Hynek Hermansky
10:55 - 11:20 Adaptation Transforms of Auto-Associative Neural Networks as Features for Speaker Verification
Samuel Thomas, Sri Harish Mallidi, Sriram Ganapathy and Hynek Hermansky
11:20 - 11:45 Bottleneck Features for Speaker Recognition
Sibel Yaman, Jason Pelecanos and Ruhi Sarikaya
11:45 - 12:10 Preliminary Investigation of Boltzmann Machine Classifiers for Speaker Recognition
Themos Stafylakis, Patrick Kenny, Mohammed Senoussaoui and Pierre Dumouchel
12:10 - 12:35 First attempt of Boltzmann Machines for Speaker Verification
Mohammed Senoussaoui, Najim Dehak, Patrick Kenny, Réda Dehak and Pierre Dumouchel

Session 5: Speaker Diarization
Tuesday 26 June 2012, Session chair: Andreas Stolcke

13:30 - 13:55 Online Two Speaker Diarization
Hagai Aronowitz, Yosef Solewicz and Orith Toledo-Ronen
13:55 - 14:20 On the use of Agglomerative and Spectral Clustering in Speaker Diarization of Meetings
Jordi Luque and Javier Hernando
14:20 - 14:45 Generalized Viterbi-based Models for Time-Series Segmentation Applied to Speaker Diarization
Itshak Lapidot and Jean-Francois Bonastre
14:45 - 15:10 A Global Optimization Framework For Speaker Diarization
Mickael Rouvier and Sylvain Meignier
15:10 - 15:35 Cisco’s Speaker Segmentation and Recognition System
Sachin Kajarekar, Aparna Khare, Matthias Paulik, Neha Agrawal, Panchi Panchapagesan, Ananth Sankar and Satish Gannu

Session 6: Speaker Recognition – Channel Robustness
Tuesday 26 June 2012, Session chair: Honza Černocký

16:00 - 16:25 Variance-Spectra based Normalization for I-vector Standard and Probabilistic Linear Discriminant Analysis
Pierre-Michel Bousquet, Anthony Larcher, Driss Matrouf, Jean-Francois Bonastre and Oldrich Plchot
16:25 - 16:50 Utterance Partitioning with Acoustic Vector Resampling for I-Vector based Speaker Verification
Wei RAO and Man-Wai MAK
16:50 - 17:15 Study on the Effects of Intrinsic Variation using i-Vectors in Text-Independent Speaker Verification
Sheng Chen, Mingxing Xu and Emlyn Pratt
17:15 - 17:40 Exploring the Impact of Advanced Front-End Processing on NIST Speaker Recognition Microphone Tasks
William Campbell, Doug Sturim, Jonas Borgstrom, Robert Dunn, Alan McCree, Tom Quatieri and Doug Reynolds
17:40 - 18:05 Linear Prediction Modulation Filtering for Speaker Recognition of Reverberant Speech
Bengt Borgstrom and Alan McCree

Session 7: Language Recognition Evaluation
Wednesday 27 June 2012, Session chair: Tomi Kinnunen

10:30 - 10:55 Evaluation of Spoken Language Recognition Technology Using Broadcast Speech: Performance and Challenges
Luis J. Rodriguez-Fuentes, Amparo Varona, Mireia Diez, Mikel Penagarikano and German Bordel
10:55 - 11:20 New Resources for Recognition of Confusable Linguistic Varieties: The LRE11 Corpus
Stephanie Strassel, Kevin Walker, Karen Jones, Dave Graff and Christopher Cieri
11:20 - 11:45 The MITLL NIST LRE 2011 Language Recognition System
Elliot Singer, Pedro Torres-Carrasquillo, Douglas Reynolds, Alan McCree, Fred Richardson, Najim Dehak and Doug Sturim
11:45 - 12:10 Description and analysis of the Brno276 system for LRE2011
Niko Brummer, Sandro Cumani, Ondrej Glembek, Martin Karafiat, Pavel Matejka, Jan Pesan, Oldrich Plchot, Mehdi Soufifar, Edward de Villiers and Jan Cernocky
12:10 - 12:35 A Linguistic Data Acquisition Front-End for Language Recognition Evaluation
Gang Liu, Chi Zhang and John Hansen

Session 8: Features for Speaker Recognition
Wednesday 27 June 2012, Session chair: John Hansen

13:30 - 13:55 Feature Extraction Using 2-D Autoregressive Models For Speaker Recognition
Sriram Ganapathy, Samuel Thomas and Hynek Hermansky
13:55 - 14:20 Regularization of All-Pole Models for Speaker Verification Under Additive Noise
Cemal Hanilci, Tomi Kinnunen, Rahim Saeidi, Jouni Pohjalainen, Paavo Alku and Figen Ertas
14:20 - 14:45 Factor Analysis of Acoustic Features using a Mixture of Probabilistic Principal Component Analyzers for robust Speaker Verification
Taufiq Hasan and John Hansen
14:45 - 15:10 Exemplar-based Sparse Representation and Sparse Discrimination for Noise Robust Speaker Identification
Rahim Saeidi, Antti Hurmalainen, Tuomas Virtanen and David A. van Leeuwen
15:10 - 15:35 On the use of Asymmetric-shaped Tapers for Speaker Verification using I-vectors
Md Jahangir Alam, Patrick Kenny and Douglas O'Shaughnessy

Session 9: Speaker Recognition Evaluation
Thursday 28 June 2012, Session chair: William Campbell

10:00 - 10:25 The Effect of Target/Non-Target Age Difference on Speaker Recognition Performance
George Doddington
10:25 - 10:50 Variational Bayes Logistic Regression as Regularized Fusion for NIST SRE 2010
Ville Hautamäki, Kong Aik Lee, Anthony Larcher, Tomi Kinnunen, Bin Ma and Haizhou Li
10:50 - 11:15 The 2011 BEST Speaker Recognition Interim Assessment
Craig Greenberg, Alvin Martin and Mark Przybocki
11:15 - 11:40 The REPERE Challenge: finding people in a multimodal context
Juliette Kahn, Olivier Galibert, Matthieu Carré, Aude Giraudel, Philippe Joly and Ludovic Quintard
11:40 - 12:05 The RATS Radio Traffic Collection System
Kevin Walker and Stephanie Strassel

Session 10: Speaker Recognition – Application
Thursday 28 June 2012, Session chair: Man-Wai Mak

13:00 - 13:25 Effects of Audio and ASR Quality on Cepstral and High-level Speaker Verification Systems
Andreas Stolcke, Martin Graciarena and Luciana Ferrer
13:25 - 13:50 Audio Context Recognition in Variable Mobile Environments from Short Segments using Speaker and Language Recognizers
Tomi Kinnunen, Rahim Saeidi, Jussi Leppaanen and Jukka P. Saarinen
13:50 - 14:15 Text Dependent Speaker Verification Using a Small Development Set
Hagai Aronowitz
14:15 - 14:40 A Unified Approach for Audio Characterization and its Application to Speaker Recognition
Luciana Ferrer, Lukas Burget, Oldrich Plchot and Nicolas Scheffer
14:40 - 15:05 Mean Shift Algorithm for Exponential Families with Applications to Speaker Clustering
Themos Stafylakis, Vassilis Katsouros, Patrick Kenny and Pierre Dumouchel

Session 11: Language Recognition – Feature, Classifier and Fusion
Thursday 28 June 2012, Session chair: Michael Wagner

15:30 - 15:55 Speaker Vectors from Subspace Gaussian Mixture Model as Complementary Features for Language Identification
Oldrich Plchot, Martin Karafiat, Niko Brummer, Ondrej Glembek, Pavel Matejka, Edward de Villiers and Jan Cernocky
15:55 - 16:20 Complementary Combination in i-Vector Level for Language Recognition
Zhi-Yi Li, Wei-Qiang Zhang, Liang He and Jia Liu
16:20 - 16:45 Bhattacharyya-based GMM-SVM System with Adaptive Relevance Factor for Pair Language Recognition
Changhuai You, Haizhou Li, Eliathamby Ambikairajah, Kong Aik Lee and Bin Ma
16:45 - 17:10 Fusing Language Information from Diverse Data Sources for Phonotactic Language Recognition
Mohamed Faouzi BenZeghiba, Jean-Luc Gauvain and Lori Lamel

Odyssey 2012 Welcome Reception
Monday 25 June 2012, 17:45 – 22:00

The welcome reception will take place at TAVERN @ THE TIGER BREWERY, immediately after the last technical session on Monday. After a short tour to the Tiger Brewery, a beer-tasting session serving seven types of beer freshly brewed at the brewery will take place alongside with a buffet dinner. Karaoke, pool table and dart board will also be available.

The Welcome Reception is included as part of the Odyssey 2012 registration. For the way back to hotels, buses are chartered to reach four MRT stations (Buona Vista, Outram Park, City Hall and Bugis). There are two departures from the brewery, the first at 21:00 and the second at 22:00.

Odyssey 2012 Social Event – Singapore Marina Sightseeing
Wednesday 27 June 2012, 15:35 - 22:00

The social event will take place at Singapore central area. The event will begin right after the last technical session on Wednesday. Buses will bring the participants to the Singapore River side from where you will board the boats to cruise along Singapore River all the way to Singapore Flyer – the world largest observation wheel. After enjoying the panoramic view over the Marina Bay, dinner will be served at the Megu Restaurant located next to the Flyer during which Jazz music will be played by Fuchsia. The tentative schedule is as follows:

15:35 End of technical session
16:00 Departure from the Matrix building in Biopolis
16:30 Arrival at the Merlion Esplanade
17:00 Board Singapore River cruise boats
17:45 Arrival at Singapore Flyer, ride the giant observation wheel
18:30 Welcome drink at Megu restaurant (within the Flyer's building), enjoy the view at the Marina Bay
19:00 Banquet at the Megu Restaurant
22:00 Departure from the Megu Restaurant, buses are chartered to reach four MRT stations ( Bugis, City Hall, Outram Park and Buona Vista)

The social event is included as part of the Odyssey 2012 registration. Accompanying persons are welcomed to attend the banquet at SGD 150.00 per person.

 

 


 
 
 
 


Copyright @ 2011-2012 COLIPS