|
|
9:00 Registration / Welcome
9:30 Keynote 1 - Noise robustness
10:30 Tea/coffee
11:00 Oral presentations
-
Multiple stream model-based
feature enhancement for noise robust speech recognition
Veronique Stouten, Hugo Van hamme, Patrick Wambacq,
ESAT, Belgium
-
Cepstral mean and
variance normalization in the model domain
Ole Morten Strand, Andreas Egeberg, NUST, Norway
-
On de-emphasizing
the spurious components in the spectral modulation
for robust speech recognition
Vivek Tyagi, Christian Wellekens, Eurecom, France
-
Towards improving
the robustness of distributed speech recognition in
packet loss
Alastair James, Ben Milner, University of East Anglia,
UK
-
A comparative study
of feature-domain error concealment techniques for
distributed speech recognition
Zheng-Hua Tan, Borge
Lindberg, Paul Dalsgaard, SMS, Denmark
1:00 Lunch
2:00 Keynote 2 - Robustness to Transmission Channel
- The DSR Approach
3:00 Tea/coffee
3:30 Poster presentations
-
A formant tracking
LP model for speech processing in car/train noise
Qin Yan, Esfandiar Zavarehei, Saeed Vaseghi, Brunel
University, UK
-
Robust SNR estimation
of noisy speech based on Gaussian mixtures modeling
on log-power domain
Tran Huy Dat, Kazuya
Takeda, Fumitada Itakura, CIAIR, Japan
-
Subspace kernel discriminant
analysis for speech recognition
Hakan Erdogan, Sabanci
University, Turkey
-
Harmonic tracking-based
short-time chirp analysis of speech signals
Marian Kepesi, Luis
Weruaga, FTW, Austria
-
Hidden Markov models
in voice activity detection
Jirí Tatarinov,
Petr Pollák, CTU, Czech Republic
-
Formant prediction
from MFCC vectors
Jonathan Darch,
Ben Milner, University of East Anglia, UK
-
Correlation based
soft-decoding for distributed speech recognition over
IP networks
Antonio Cardenal-López,
Carmen García-Mateo, University of Vigo,
Spain
-
Statistical-based
reconstruction methods for speech recognition in IP
networks
Angel Manuel Gómez,
Antonio Miguel Peinado, Victoria Sánchez,
Univ. Granada, Spain
-
Packet loss modelling
for distributed speech recognition
Ben Milner, Alastair
James, University of East Anglia, UK
5:00 End of day 1
7:30 Conference Dinner
9:00 Keynote 3 - Multimodalies,
10:00 Tea/coffee
10:20 Oral presentations
-
Early error
detection on word level
Gabriel Skantze,
Jens Edlund, KTH, Sweden
-
Making an information
state-based spoken dialogue system less fragile
William Black,
UMIST, UK
-
Audio-visual
speech recognition using new lip features extracted
from side-face images
Tomoaki Yoshinaga,
Koji Iwano, Sadaoki Furui, Tokyo Institute, Japan
-
Benefits of
disfluency detection in spontaneous speech recognition
Frederik Stouten,
Jean-Pierre Martens, ELIS, Belgium
12:00 Lunch
1:00 Keynote 4 - Robust
conversational system design
-
Lou Boves, University of Nijmegen,
The Netherlands
2:00 Tea/coffee
2:00 Poster presentations
-
Task independent
speech verification using SB-MVE trained phone models
Magne H. Johnsen, Svein
G. Pettersen, Tor A. Myrvoll, NTNU, Norway
-
Robust interpretation
in the Higgins spoken dialogue system
Gabriel Skantze,
Jens Edlund, KTH, Sweden
-
Flexible knowledge
representation for robust dialogue management
Melita Hajdinjak,
France Mihelic, University of Ljubljana, Slovenia
-
Dynamic choice
of robust strategies
Bryan McEleney,
University College Dublin, Ireland
-
A robust input
interface in the scope of the project interactive
home of the future
João
P. Neto, INESC ID Lisboa, Portugal
-
Multimodal speaker
identification using adaptive decision fusion with
reliability weighted summation
Engin Erzin,
Yucel Yemez, A. Murat Tekalp, Koc University, Turkey
-
Increasing interaction
robustness of speech-enabled mobile applications by
enhancing speech output with non-speech sound
Peter Fröhlich,
FTW, Austria
-
Combined use
of close-talk and throat microphones for improved
speech recognition under non-stationary background
noise
Stéphane
Dupont, Christophe Ris, Damien Bachelart, Multitel,
Belgium
-
Adding robustness
to language models for spontaneous speech recognition
Jacques Duchateau,
Tom Laureys, Patrick Wambacq, ESAT, Belgium
-
Crosslingual
transfer of source acoustic models to two different
target languages
Andrej Zgank,
Zdravko Kacic, Klara Vicsi Gyorgy Szaszak, Frank
Diehl, Jozef Juhar, Slavomir Lihan, University of
Maribor, Slovenia
-
Automatic recognition
of spontaneously dictated medical records for Norwegian
Bojana Gajic,
Vidr Markhus, Svein Gunnar Pettersen, NUST, Norway
3:00 Panel discussion
3:30 END - Bus leaves
for Airport
|