2002 International Workshop

Speech and Computer

SPECOM'2002

PROGRAM

St. Petersburg, Russia

2 - 5 September 2002

Organized by

St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences (SPIIRAS)

Sponsored by

The Ministry of Industry, Science and Technologies

GENERAL CHAIRMAN'S MESSAGE

Distinguished quests, ladies and gentlemen, dear colleagues!  It is great pleasure for me as General Chairman to welcome you to the 7-th International Workshop on Speech and Computer in St. Petersburg. 

The SPECOM - series established by St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences was intended to discuss the most important fundamental aspects of man-computer speech interaction and more perspective application of this research. We hope that Workshop will be useful for your future work.

We would like to convey our best wishes to all the participants and guests.

I wish you success.

Rafael Yusupov

Chairman, SPECOM’2002 

ORGANIZING COMMITTEE CHAIR'S MESSAGE

Distinguished guests, ladies and gentlemen, dear colleagues! It is a great pleasure for me as Chairman of the Organizing Committee to welcome you to the 7-th SPECOM Workshop.

We are pleased that so many outstanding researchers from Russia and from some other countries have come to attend this Workshop. We would like to convey our best wishes to all the participants and guests.

Our research are developed impetuously. But there are some challenges and obstacles preventing successful application of speech systems. So let us solve this problem together!

I wish you success here  and in your research. Thank you.

Yuri Kosarev

SPECOM’2002 Organizing Committee Chair

TABLE OF CONTENTS

	SURVEY LECTURES:
SL.1	Robustness in Speech and Text Processing Systems (Linguistic Automaton) .............. X.Piotrowska, R. Piotrowski, Y.Romanov Herzen State Pedagogical University of Russia, Russia	1
SL.2	Some aspects of Robust Speech Understanding ............................................... Yuri Kosarev St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences, Russia	3

Session 1:	SPEECH UNDERSTANDING. DIALOG MODELS. LANGUAGE MODELS: Chair: Jean Caelen
1.1	Neural Network and Information Theory in Automatic Speech Understanding.......... Salma Jamoussi, Kamel Smaïli and Jean-Paul Haton LORIA/INRIA-Lorraine, FRANCE	9
1.2	Robust Speech Understanding for Voice Control System ......................................... Yuri Kosarev, Izolda Lee, Andrey Ronzhin, Alexey Karpov (1), Jesus Savage (2), Fred Haritatos (3) (1) (1) St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences, Russia (2) Intelligent Interfaces Laboratory University of Mexico, Mexico (3) Air Force Research Laboratory NY, USA	13
1.3	Design of Semantic Language Model for Czech Train Timetable Inquiry Processing... Libor Jelinek University of West Bohemia, Department of Cybernetics, Czech Republic	19
1.4	Lattice Rescoring in Czech LVCSR System Using Linguistic Knowledge ................... Pavel Ircing and Josef Psutka University of West Bohemia, Department of Cybernetics, Czech Republic	23
1.5	About using non-informational functions in models of speech communication............ Soloviev A.N., Victorova K.O., Razumikhin D.V. Auditech, Russia	27

Session 2:	SPEECH RECOGNITION: Chair: Roger  Marshall
2.1	Speech-To-Text Systems (STS): Issues in Training, Multi-Lingual Transcription and Speaker’s Linguistic Identity.................................................................................. R. Marshall Math & Computer Science Dept., Bridgewater State College , USA	31
2.2	Recognition and Rejection Performance in Wordspotting Using Hidden Markov Modeling Techniques ............................................................................................ Yassine Ben Ayed ^1, Dominique Fohr^1, Jean Paul Haton^1 , Gérard Chollet^2 (1) LORIA-CNRS/ INRIA Lorraine, France (2) ENST, CNRS-LTCI, France	35
2.3	Support Vector Machines for Keyword Spotting...................................................... Yassine Ben Ayed ^1, Dominique Fohr^1, Jean Paul Haton^1 , Gérard Chollet^2 (1) LORIA-CNRS/ INRIA Lorraine, France (2) ENST, CNRS-LTCI, France	39
2.4	Dynamic Topic Identification: Introduction of Trigger pairs in the Cache Model....... Brigitte Bigi, Salma Jamoussi, Kamel Smaïli LORIA/INRIA Lorraine, FRANCE	43
2.5	Continuous Speech Recognition Method suitable for Robust Speech Understanding.... Andrey Ronzhin, Yuri Kosarev, Izolda Lee, Alexey Karpov St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences, Russia	47
2.6	A Study of the Acoustic Model Choice for Russian Speech Recognition...................... Chuchupal Vladimir, Makovkin Konstantin, Gorokhovsky Konstantin, Chichagov Alexander Computer Center, Russian Academy of Science, RUSSIA	53
2.7	VQ and GMM Combination for Text-Independent  Speaker Recognition on Telephone Channel................................................................................................................. V.I.Galunov, S.N.Gramnitski, N.A.Romashov, A.B.Viktorov Auditech, Russia	57
Session 3:	SIGNAL PROCESSING.  PRONUNCIATION MODELING: Chair: Nobuhiro Miki
3.1	High Speed Computation of Transfer Function of 3-D Vocal Tract............................ Nobuhiro Miki *, Kohji Sasaki**, and Hiroki Matsuzaki*** * Department of Media Architecture Future University - Hakodate, Japan ** Graduate School of Engineering, Hokkaido University Sapporo, Japan *** Department of Electronic Information Engineering Hokkai Gauen University, Japan	61
3.2	Voice Character Transformation............................................................................ Michael Savic Rensselaer Polytechnic Institute, Speech and Signal Research Group,Troy, USA	65
3.3	The Cancellation of Loudspeaker Interference......................................................... Jan Dohnal New Technologies Research Centre, Czech Republic	69
3.4	The Speech Signal Analysis and Problems of Speech Understanding.......................... Vladimir Bondarenko, Roman Mescheriakov, Vladislav Kotsubinski, Andrew Ponomarev Tomsk State University of the Control system and Radioelectronics, Russia	73

Session 4:	ACOUSTICAL MODELING. SPEECH SYNTHESIS:  Chair: Valery Galunov
4.1	The Acoustical Aspect of Vocal-Speech Communication During the Early Childhood.. Galunov V., Lyakso E., Chelibanova O, Petrikova N. The Department of Higher Nervous Activity and Psychophysiology, St-Petersburg State University, Russia	75
4.2	On Virtual and Real Pauses at Syntactic Boundaries in Russian................................ Nina B.Volskaya St.Petersburg State University, Department of Phonetics, Russia	83
4.3	Modelling of Sonants............................................................................................. Yakushev Denis Igorevich SPb State electrotechnical university, Russia	87
4.4	About Voice Intonation Dimension.......................................................................... A.G.Boyarov - Moscow state institute of electronics and mathematics, Russia S.N.Krinov - Department of Linguistics of Moscow pedagogical university (Moscow state regional university), Russia	91
4.5	Audio Indexing System for Segmentation, Classification, Attributing and Indexing of Broadcast Stream.................................................................................................. Sergei N. Gramnitski, Alexandr L. Maiboroda, Victor V. Redkov, Mikhael A. Sall, Anatoli I. Tikhotski, Andrei B. Victorov LG Soft Lab, Russia	95
4.6	Statistical Approach to Creating Phonetic Transcription.......................................... A.V. Agranovsky, M.Y. Zulkarneyev, D.A. Lednov GP KB “Spetsvuzavtomatika”, Russia	99
4.7	Speech Synthesis in Dialogue Systems..................................................................... Vladimir Bondarenko, Roman Mescheriakov, Vladislav Kotsubinski Tomsk State University of the Control system and Radioelectronics, Russia	103

Session 5:	LANGUAGE PROCESSING. LANGUAGE RESOURCES:  Chair: Rodmonga Potapova
5.1	Foreign Language Database of the Interactive Expert System “Dialect” for Forensic Speaker Identification Used in Russia........................................................................ R.K. Potapova*, V.V. Potapov** *Moscow State Linguistic University, Russia **Moscow State University, Russia	107
5.2	Wideband speech database for Russian.................................................................. Galunov V.I., Kochanina J.L., Ostroukhov A.V., Solovyev A.N., Evdomakha A.V., Razumikhin D.V. (Auditech, Russia),  Tropf H., Hoege H. (Siemens AG, Germany)	113
5.3	Promising NLP-Technologies.............................................................................. Natalia Zaitseva Herzen State Pedagogical University of Russia, Russia	117
5.4	Predicting of the Statements Motive In Telecommunications Dialogue Systems....... Max V. Klimovich 1) , Roman Mescheriakov 2) 1) "Surguttel",Surgut State University, Russia, 2) Tomsk State University of the Control Systems and Electronics,  Russia	121

Proceedings copy is $ 15.

CONTACT ADDRESS

All correspondence regarding the workshop should be addressed to:

Dr. Yu. Kosarev

St. Petersburg Institute for Informatics and Automation,

39, 14^th line, St. Petersburg, 199178, RUSSIA

Tel.:   812 328 7081

Fax :   812 328 4450

E-mail: specom@mail.iias.spb.su

http://www.spiiras.nw.ru/speech