Odyssey 2014

The Speaker and Language Recognition Workshop

Odyssey 2014

Odyssey 2014: The Speaker and Language Recognition Workshop was hosted by the School of Computing of University of Eastern Finland (UEF) in Joensuu, Finland, on June 16-19, 2014. Odyssey 2014 is an ISCA tutorial and research workshop held in cooperation with the ISCA Speaker and Language Characterization special interest group. The need for fast, efficient, accurate, and robust means of recognizing people and languages is of growing importance for commercial, forensic, and government applications. The aim of this workshop is to continue to foster interactions among researchers in speaker and language recognition as the successor of previous successful events held in Martigny (1994), Avignon (1998), Crete (2001), Toledo (2004), San Juan (2006), Stellenbosch (2008), Brno (2010) and Singapore (2012).

Website: http://cs.uef.fi/odyssey2014/


Keynotes


Opening & Closing

0:16:31

Opening

Tomi Kinnunen

0:15:07

Closing, Awards & Odyssey 2016

Tomi Kinnunen, Pasi Fränti, Mikel Peñagarikano


Calibration, Evaluation & Forensics

0:16:58

Effects of the New Testing Paradigm of the 2012 NIST Speaker Recognition Evaluation

Alvin F. Martin, Craig S Greenberg, Vincent M. Stanford, John M. Howard, George R. Doddington and John J. Godfrey

0:33:18

NFI-FRITS: A forensic speaker recognition database and some first experiments

David van der Vloed, Jos Bouten and David Van Leeuwen

0:15:32

A comparison of linear and non-linear calibrations for speaker recognition

Niko Brummer, David van Leeuwen and Albert Swart


0:21:46

Trial-based Calibration for Speaker Recognition in Unseen Conditions

Mitchell Mclaren, Aaron Lawson, Luciana Ferrer, Nicolas Scheffer and Yun Lei


Speaker Modeling I

0:25:13

What are we missing with i-vectors? A perceptual analysis of i-vector-based falsely accepted trials

Joaquin Gonzalez-Rodriguez, Juana Gil, Rubén Pérez and Javier Franco-Pedroso

0:23:18

Exploring some limits of Gaussian PLDA modeling for i-vector distributions

Pierre-Michel Bousquet, Jean-François Bonastre and Driss Matrouf


0:25:26

GMM Weights Adaptation Based on Subspace Approaches for Speaker Verification

Najim Dehak, Oldrich Plchot, Mohamad Hasan Bahari, Lukas Burget, Hugo Van Hamme and Reda Dehak


Language Recognition

0:15:47

NIST Language Recognition Evaluation – Past and Future

Alvin F. Martin, Craig S. Greenberg, John M. Howard, George R. Doddington and John J. Godfrey

0:21:56

Robust Language Recognition Based on Diverse Features

Qian Zhang, Gang Liu and John Hansen

0:29:27

Speaker-basis Accent Clustering Using Invariant Structure Analysis and the Speech Accent Archive

Nobuaki Minematsu, Shun Kasahara, Takehiko Makino, Daisuke Saito and Keikichi Hirose



Speaker Diarization

0:27:14

Telephone Conversation Speaker Diarization Using Mealy-HMMs

Itshak Lapidot, Jean-Francois Bonastre and Samy Bengio

0:25:09

Person Instance Graphs for Named Speaker Identification in TV Broadcast

Hervé Bredin, Antoine Laurent, Achintya Sarkar, Viet-Bac Le, Sophie Rosset and Claude Barras

0:21:02

Recent Improvements on ILP-based Clustering for Broadcast News Speaker Diarization

Grégor Dupuy, Sylvain Meignier, Paul Deléglise and Yannick Estève


0:26:15

Modeling Overlapping Speech using Vector Taylor Series

Pranay Dighe, Marc Ferras and Herve Bourlard


Text-dependent Speaker Recognition

0:21:55

Joint Factor Analysis for Text-Dependent Speaker Verification

Patrick Kenny, Themos Stafylakis, Alam Jahangir, Pierre Ouellet and Marcel Kockmann

0:26:08

Short-Duration Speaker Modelling with Phone Adaptive Training

Giovanni Soldi, Simon Bozonnet, Federico Alegre, Christophe Beaugeant and Nicholas Evans

0:26:45

Text-Dependent Speaker Verification System in VHF Communication Channel

Changhuai You, Kong Aik Lee, Bin Ma and Haizhou Li



Nist I-Vector Special Session

0:21:44

The NIST 2014 Speaker Recognition i-vector Machine Learning Challenge

Craig Greenberg, Désiré Bansé, George Doddington, Daniel Garcia-Romero, John Godfrey, Tomi Kinnunen, Alvin Martin, Alan McCree, Mark Przybocki and Douglas Reynolds

0:21:48

STC Speaker Recognition System for the NIST i-Vector Challenge

Sergey Novoselov, Timur Pekhovsky and Konstantin Simonchik

0:10:54

Incorporating Duration Information into I-Vector-Based Speaker Recognition Systems

Bostjan Vesnicer, Jerneja Zganec-Gros, Simon Dobrisek and Vitomir Struc


0:17:08

Linearly Constrained Minimum Variance for Robust I-vector Based Speaker Recognition

Abbas Khosravani and Mohammad Mahdi Homayounpour

0:18:41

Hierarchical speaker clustering methods for the NIST i-vector Challenge

Elie Khoury, Laurent El Shafey, Marc Ferras and Sebastien Marcel


Speaker Modeling II

0:23:08

Unsupervised Domain Adaptation for I-Vector Speaker Recognition

Daniel Garcia-Romero, Alan McCree, Stephen Shum, Niko Brummer and Carlos Vaquero

0:25:55

Unsupervised Clustering Approaches for Domain Adaptation in Speaker Recognition Systems

Stephen Shum, Douglas Reynolds, Daniel Garcia-Romero and Alan McCree

0:18:09

Generative pairwise models for speaker recognition

Sandro Cumani and Pietro Laface



Neural Nets for Speaker and Language Modeling

0:16:37

Application of Convolutional Neural Networks to Language Identification in Noisy Conditions

Yun Lei, Luciana Ferrer, Aaron Lawson, Mitchell McLaren and Nicolas Scheffer

0:22:57

Deep Neural Networks for extracting Baum-Welch statistics for Speaker Recognition

Patrick Kenny, Vishwa Gupta, Themos Stafylakis, Pierre Ouellet and Jahangir Alam

0:26:41

Neural Network Bottleneck Features for Language Identification

Pavel Matejka, Le Zhang, Tim Ng, Sri Harish Mallidi, Ondrej Glembek, Jeff Ma and Bing Zhang