1 . Editorial

Dear members,

The outstanding phonetician Nick Clement passed away end of August. His French colleagues sent me his obituary. Our speech community will miss this excellent colleague and collaborator.

After the successful Interspeech in Brighton, it is time to think about your contributions to the next conferences and more particularly to Interspeech 2010 to be held in Chiba, Japan.

I met several of you in Brighton and I thank all of you who told me how important for them is this monthly newsletter. In order to improve the quality of information you will find in it, I ask all conference, workshop, special issue organizers to send me as soon as they make their decision about the organization to send me their Calls for papers and a description of its objectives.

Many speech scientists found new jobs using our list of job openings. If you are in a company or University planning to hire new scientists, professors, post docs, PhD students, please ask the persons responsible to contact me and to advertise in ISCApad.

If you recently wrote a book devoted to Speech science or technology, send me its identification with some description of its content. It will be advertised in ISCApad.

I try to gather a maximum of news about the speech community: please help me to do better with your personal contribution. Thanks for all members.

Prof. em. Chris Wellekens

Institut Eurecom

Sophia Antipolis
France

public@isca-speech.org

2 . ISCA News

2-1 . Nick Clement's obituary

G. Nick Clements’ death on Aug. 30, 2009 shocked everyone who was fortunate enough to know him. We have lost a great friend and one of the most gifted and productive phonologists of the last decades.

In order to keep Nick’s fertile spirit alive, the Laboratoire de Phonétique et Phonologie (LPP) has set up a web page where colleagues and friends can share memories and thoughts about Nick. This page will be updated frequently. It is accessible at http://lpp.univ-paris3.fr/equipe/nick_clements/remembering-nick-clements.html.

If you wish to contribute, please send text, images or videos to remember.nickclements@gmail.com

We will gather to remember Nick on Friday October 9^th, 2009 from 6 pm at the LPP (19 rue des Bernardins, 75005 Paris). If you wish to share your memories of Nick and celebrate his life and what he brought us, you are invited to join us.

For organizational reasons, we would appreciate to know in advance if you will join us. Please email remember.nickclements@gmail.com

Nick’s colleagues of the LPP in Paris

An obituary by the members of the LPP*

G. Nick Clements was born on Oct. 5, 1940, in Cincinnati, Ohio, and attended Moses Brown School in Providence. In 1962, he graduated with high honors from Yale University, majoring in fine arts and was elected to Phi Beta Kappa, which celebrates the most outstanding students of arts and sciences at America’s leading colleges and universities. After a year in Nashville as a classical music DJ, he served in the Army Signal Corps for two years, stationed in Germany. Following his service, he lived in Spain for several years, painting, studying art and writing for an English language periodical. In 1968, he received a certificat from the Centre de Linguistique Quantitative, Faculté des Sciences, Université de Paris. From 1971 to 1973, he was adjunct professor of American English at the University of Paris 8. In 1973, he received his Ph.D. in linguistics from the School of African and Oriental Studies, University of London, defending a thesis on the Ewe language based on fieldwork in Ghana. He was a visiting scientist and lecturer at the department of foreign languages and linguistics at MIT (1973-1975), and held appointments as Assistant Professor and Associate Professor at Harvard (1975-1982). In 1982, he moved to Cornell University, where he was Professor of linguistics and Director of the Phonetics Laboratory. In 1992, he came to Paris, where he became Director of Research at the Centre National de la Recherche Scientifique (CNRS) and worked in the Laboratory of Phonetics and Phonology (UMR7018). He held this position until 2008, when he was elected Professor Emeritus. Clements was also an invited professor in various prestigious universities across the world, in Europe, USA, India, Australia, etc. He was also very active in the academic world. In the last three years, he organized two widely attended international conferences at the Sorbonne University: one on “The phonetic bases of distinctive features” in 2006 and one on “Where do phonological features come from?” in 2007.

Clements’ research interests were wide-ranging and he made outstanding contributions in phonology and phonetics- phonology interface. He is best known for his groundbreaking work on syllable and feature theory and his pioneering work on the phonological systems of various African languages, including tonal and vowel-harmony systems. His recent cross-linguistic studies on phonological units have contributed in designing and developing theories and models on phonological representations and have led to a better understanding of the role of features in speech sound inventories. A characteristic feature of Clements’ works is his rigorous scientific method and his unusual gift for finding the most convincing argumentations and drawing the clearest and most synthesized conclusions. Clements was not only an excellent connoisseur of the field, not only an expert on the language or languages studied, but also an outstanding theoretician and a highly trained phonetician. He left behind for us tremendous work in the areas of phonology and phonetics. He wrote and co-authored five books and nearly 100 articles, including journal articles, book and encyclopedia chapters, conference and working papers. He was productive until the end of his life, with some major contributions still to appear.

Nick Clements had several passions outside the field of linguistics. He was a music lover and was particularly knowledgeable about jazz music. He played keyboard in a jazz workshop at a club in Paris in the last year of his life. He was also a passionate traveler and visited many parts of the world in the five continents. He traveled for both work and pleasure, and was fluent in several languages. But the number one of his passions was his family: his wife and colleague, Dr. Annie Rialland, his children, William and Célia, his brother, sisters and their families.

G. Nick Clements was a great linguist, endowed with an outstanding ability to listen, to guide, to inspire reflections, and to stimulate brainstorming and creative thinking. He was also gifted with noble human qualities: kind, compassionate, generous, and humble. He will forever be remembered fondly for that and much more.

Rachid Ridouane and all his colleagues

of the Phonetics and Phonology Laboratory in Paris.

2-2 . Interspeech proceedings indexed by Thomson Reuters (ISI)

In the closing ceremony of Interspeech 2009, the ISCA president announced that she had finally received the acceptance from Thomson Reuters (ISI) to index Interspeech Proceedings of 2006 and 2007. The day after, she received the same answer from the Engineering Index of Elsevier. She will now start the process of indexing the proceedings of 2008 and 2009 in these two databases and continue the efforts concerning other major citation indexes.

2-3 . Message to students

Dear students,

The International Speech Communication Association ISCA is now opening an online system to build a database of résumés of researchers/students working in the various fields of speech communication.
The goal of this service is to build a centralized place where many interested employers/corporations can access and search for potential candidates.

Please be advised that the posting service will be updated at 4 month intervals. Next switch will be mid October 2009.

We encourage all of you to upload an updated version of your résumé to: http://www.isca-speech.org/resumes/ and wish you good luck with a fruitful career.

Professor Helen Meng

3 . Future ISCA Conferences and Workshops (ITRW)

3-1 . (2010-09-26) CfP INTERSPEECH 2010 Chiba Japan

FIRST CALL FOR PAPERS

INTERSPEECH is the world's largest and most comprehensive conference on issues surrounding the science and technology of spoken language processing both in humans and in machines. It is our great pleasure to host INTERSPEECH 2010 in Japan, the birthplace of ICSLP, which has held two ICSLPs, in Kobe and Yokohama, in the past. The theme of INTERSPEECH 2010 is "Spoken Language Processing for All Ages, Health Conditions, Native Languages and Environments". INTERSPEECH 2010 emphasizes an interdisciplinary approach covering all aspects of speech science and technology spanning the basic theories to applications. Besides regular oral and poster sessions, plenary talks by internationally renowned experts, tutorials, exhibits, and special sessions are planned. We invite you to submit original papers in any related area, including but not limited to:

HUMAN SPEECH PRODUCTION, PERCEPTION AND COMMUNICATION

* Human speech production * Human speech and sound perception * Linguistics, phonology and phonetics * Discourse and dialogue * Prosody (e.g. production, perception, prosodic structure, modeling) * Paralinguistic and nonlinguistic cues (e.g. emotion and expression) * Physiology and pathology of spoken language * Spoken language acquisition, development and learning * Speech and other modalities (e.g. facial expression, gesture)

SPEECH AND LANGUAGE TECHNOLOGY

* Speech analysis and representation * Speech segmentation * Audio segmentation and classification * Speaker turn detection * Speech enhancement * Speech coding and transmission * Voice conversion * Speech synthesis and spoken language generation * Automatic speech recognition * Spoken language understanding * Language and dialect identification * Cross-lingual and multi-lingual speech processing * Multimodal/multimedia signal processing (including sign languages) * Speaker characterization and recognition * Signal processing for music and song * Spoken language technology for prosthesis, rehabilitation, wellness and welfare

SPOKEN LANGUAGE SYSTEMS AND APPLICATIONS

* Spoken dialogue systems * Systems for information extraction/retrieval * Systems for spoken language translation * Applications for aged and handicapped persons * Applications for learning and education * Other applications

RESOURCES, STANDARDIZATION AND EVALUATION

* Spoken language resources and annotation * Evaluation and standardization of spoken language systems

PAPER SUBMISSION

Papers for the INTERSPEECH 2010 proceedings should be up to four pages in length and conform to the format given in the paper preparation guidelines and author kits which will be available on the INTERSPEECH 2010 website along with the Final Call for Papers. Optionally, authors may submit additional files, such as multimedia files, to be included on the Proceedings CD-ROM. Authors shall also declare that their contributions are original and not being submitted for publication elsewhere (e.g. another conference, workshop, or journal). Papers must be submitted via the on-line paper submission system, which will open early in 2010. The deadline for submitting a paper is 30 April 2010. This date will not be extended. Inquiries regarding paper submissions should be directed via email to submission@interspeech2010.org.

LANGUAGE

The working language of the conference is English.

IMPORTANT DATES

Paper submission deadline: 30 April 2010

Notification of acceptance or rejection: 2 July 2010

Camera-ready paper due: 9 July 2010

Early registration deadline: 28 July 2010

Conference dates: 26-30 September 2010

WEBSITE & MAIL

http://www.interspeech2010.org/

mail: office@interspeech2010.org

VENUE

Makuhari Messe International Conference Hall Nakase 2-1, Mihama-ku, Chiba-city Chiba 261-0023 Japan http://www.m-messe.co.jp/en/access/index.html

3-2 . (2011-08-27) INTERSPEECH 2011 Florence Italy

Interspeech 2011

Palazzo dei Congressi, Italy, August 27-31, 2011.

Organizing committee

Piero Cosi (General Chair),

Renato di Mori (General Co-Chair),

Claudia Manfredi (Local Chair),

Roberto Pieraccini (Technical Program Chair),

Maurizio Omologo (Tutorials),

Giuseppe Riccardi (Plenary Sessions).

More information www.interspeech2011.org

4 . Workshops and conferences supported (but not organized) by ISCA

4-1 . (2009-11-05) Workshop on Child, Computer and Interaction

Call for Papers

The Workshop on Child, Computer and Interaction (wocci2009.fbk.eu) will be held in Boston on November 5th, 2009.

For registration visit

http://icmi2009.acm.org

The Workshop is a satellite event of the Eleventh International

Conference on Multi-modal Interfaces, this year jointly with Machine Learning Multimodal

Interaction (ICMI-MLMI 2009) that will take place in the same venue

November 2-4, 2009.

This Workshop aims at bringing together researchers and practitioners from

universities and industry working in all aspects of child-machine interaction including

computer, robotics and multi-modal interfaces. Children are special both at the

acoustic/linguistic level but also at the interaction level. The Workshop provides a

unique opportunity for bringing together different research communities from

cognitive science, robotics, speech processing, linguistics and application areas

such as medical and education. Various state-of-the-art components can be

presented here as key components for next generation child centred computer

interaction. Technological advances are increasingly necessary in a world where

education and health pose growing challenges to the core well-being of our

societies. Noticeable examples are remedial treatments for children with or without

disabilities and individualised attention. The Workshop should serve for presenting

recent advancements in core technologies as well as experimental systems and

prototypes.

Technical Scope

The technical scope of the workshop includes, but it is not limited to:

● Speech Interfaces: acoustic and linguistic analysis of children's speech, discourse

analysis of spoken language in child-machine interaction, age-dependent

characteristics of spoken language, automatic speech recognition for children and

spoken dialogue systems

● Multi-modality and Robotics: multi-modal child-machine interaction, multi-modal

input and output interfaces, including robotic interfaces, intrusive, non-intrusive

devices for environmental data processing, pen or gesture/visual interfaces

● User Modelling: user modelling and adaptation, usability studies accounting for

age preferences in child-machine interaction

● Cognitive Models: internal learning models, personality types, user-centred and

participatory design

● Application Areas: training systems, educational software, gaming interfaces,

medical conditions and diagnostic tools

Paper submission

Authors are invited to submit papers in any technical areas relevant to the workshop.

The technical committee will select papers for oral/poster presentations.

Demonstrations are especially welcome. Instructions for paper submission are

available at the Workshop website. An electronic version of Workshop proceedings will be published by ACM

Chairs

Kay Berkling

(Inline Internet Online GmbH,

Germany)

Diego Giuliani

(FBK, Italy)

Shrikanth Narayanan

(Univ. Sourthern California, USA

4-2 . (2009-12-13) ASRU 2009

IEEE ASRU2009
Automatic Speech Recognition and Understanding Workshop
Merano, Italy December 13-17, 2009
http://www.asru2009.org/

We are happy to inform you that IEEE ASRU 2009 registrations are now available at: http://www.asru2009.org/index.php/registration  Please note the   1) At least one author/paper needs to register 2) We have a limited number of slots for registrations and priority will be given to authors of accepted papers until October 7 2009.     AFTER that registrations will accepted on a first-come first-serve basis

The eleventh biannual IEEE workshop on Automatic Speech Recognition
and Understanding (ASRU) will be held on December 13-17, 2009.
The ASRU workshops have a tradition of bringing together
researchers from academia and industry in an intimate and
collegial setting to discuss problems of common interest in
automatic speech recognition and understanding.

Workshop topics

• automatic speech recognition and understanding
• human speech recognition and understanding
• speech to text systems
• spoken dialog systems
• multilingual language processing
• robustness in ASR
• spoken document retrieval
• speech-to-speech translation
• spontaneous speech processing
• speech summarization
• new applications of ASR.

The workshop program will consist of invited lectures, oral
and poster presentations, and panel discussions. Prospective
authors are invited to submit full-length, 4-6 page papers,
including figures and references, to the ASRU 2009 website
http://www.asru2009.org/.
All papers will be handled and reviewed electronically.
The website will provide you with further details. Please note
that the submission dates for papers are strict deadlines.

IMPORTANT DATES

Paper submission deadline         July 15, 2009
Paper notification of acceptance     September 3, 2009
Demo session proposal deadline        September 24, 2009
Early registration deadline        October 7, 2009
Workshop                 December 13-17, 2009

Please note that the number of attendees will be limited and
priority will be given to paper presenters. Registration will
be handled via the ASRU 2009 website,
http://www.asru2009.org/, where more information on the workshop
will be available.

General Chairs
    Giuseppe Riccardi, U. Trento, Italy
    Renato De Mori, U. Avignon, France

Technical Chairs
    Jeff Bilmes, U. Washington, USA
    Pascale Fung, HKUST, Hong Kong China
    Shri Narayanan, USC, USA
    Tanja Schultz, U. Karlsruhe, Germany

Panel Chairs
    Alex Acero, Microsoft, USA
    Mazin Gilbert, AT&T, USA

Demo Chairs
    Alan Black, CMU, USA
    Piero Cosi, CNR, Italy

Publicity Chairs
    Dilek Hakkani-Tür, ICSI, USA
    Isabel Trancoso, INESC -ID/IST, Portugal

Publication Chair
    Giuseppe di Fabbrizio, AT&T, USA

Local Chair
    Maurizio Omologo, FBK-irst, Italy

4-3 . (2009-12-14) 6th International Workshop on Models and Analysis of Vocal Emissions for Biomedical Applications MAVEBA 2009

University degli Studi di Firenze Italy

Department of Electronics and Telecommunications

6th International Workshop

Models and Analysis of Vocal Emissions for Biomedical

Applications

MAVEBA 2009

December 14 - 16, 2009

Firenze, Italy

http://maveba.det.unifi.it

Speech is the primary means of communication among humans, and results from

complex interaction among vocal folds vibration at the larynx and voluntary articulators

movements (i.e. mouth tongue, jaw, etc.). However, only recently has research

focussed on biomedical applications. Since 1999, the MAVEBA Workshop is

organised every two years, aiming to stimulate contacts between specialists active in

clinical, research and industrial developments in the area of voice signal and images

analysis for biomedical applications. This sixth Workshop will offer the participants

an interdisciplinary platform for presenting and discussing new knowledge in the field

of models, analysis and classification of voice signals and images, as far as both

adults, singing and children voices are concerned. Modelling the normal and

pathological voice source, analysis of healthy and pathological voices, are among the

main fields of research. The aim is that of extracting the main voice characteristics,

together with their deviation from “healthy conditions”, ranging from fundamental

research to all kinds of biomedical applications and related established and advanced

technologies.

SCIENTIFIC PROGRAM

linear and non-linear models of voice

signals;

physical and mechanical models;

aids for disabled;

measurement devices (signal and image); prostheses;

robust techniques for voice and glottal

analysis in time, frequency, cepstral,

wavelet domain;

neural networks, artificial intelligence and

other advanced methods for pathology

classification;

linguistic and clinical phonetics; new-born infant cry analysis;

neurological dysfunction; multiparametric/multimodal analysis;

imaging techniques (laryngography,

videokymography, fMRI);

voice enhancement;

protocols and database design;

Industrial applications in the biomedical

field;

singing voice;

speech/hearing interactions;

DEADLINES

30 May 2009 Submission of extended abstracts (1-2 pages, 1 column)

/special session proposal

30 July 2009 Notification of paper acceptance

30 September 2009 Final full paper submission (4 pages, 2 columns, pdf

format) and early registration

14-16 December 2009 Conference venue

4-4 . (2010-05-03) Workshop on Spoken Languages Technologies for Under-Resourced Languages (SLTU'10)

Workshop on Spoken Languages Technologies for Under-Resourced Languages

(SLTU’10)

*The second International Workshop on Spoken Languages Technologies for

Under-resourced languages (SLTU’10) will be held at Universiti Sains Malaysia

(USM), Penang, Malaysia, May 3 to May 5, 2010.* Workshop

supported by ISCA, AFCP and CNRS.

The first workshop on Spoken Languages Technologies for Under-Resourced

Languages was organized in Hanoi, Vietnam, in 2008 by Multimedia,

Information, Communication and Applications (MICA) research center in

Vietnam and /Laboratoire d’Informatique de Grenoble/ (LIG) in France.

This first workshop gathered 40 participants during two days.

For 2010, we intend to attract more participants, especially from the

local regional zone (Malaysia, Indonesia, Singapore, Thailand,

Australia, ...). The workshop will take place inside USM in Penang,

Malaysia. SLTU research workshop will focus on spoken language

processing for under-resourced languages and aims at gathering

researchers working on:

* ASR, synthesis and translation for under-resourced languages

* portability issues

* multilingual spoken language processing

* fast resources acquisition (speech, text, lexicons, parallel corpora)

* spoken language processing for languages with rich morphology

* spoken language processing for languages without separators

* spoken language processing for languages without writing system

* NLP for rare or endangered languages

* …

*Important dates*

* Paper submission: December 15, 2009

* Notification of Paper Acceptance: February 15, 2010

* Author Registration Deadline: March 1, 2010

*Workshop Web site*

* *

http://www.mica.edu.vn/sltu-2010

*Workshop Chairs*

Laurent Besacier

Eric Castelli

Dr. Chan Huah Yong

4-5 . (2010-05-19) CfP LREC 2010 - 7th Conference on Language Resources and Evaluation

LREC 2010 - 7th Conference on Language Resources and Evaluation

FIRST ANNOUNCEMENT AND CALL FOR PAPERS

MEDITERRANEAN CONFERENCE CENTRE, VALLETTA - MALTA

MAIN CONFERENCE: 19-20-21 MAY 2010

WORKSHOPS and TUTORIALS: 17-18 MAY and 22-23 MAY 2010

Conference web site: http://www.lrec-conf.org/lrec2010/

The seventh international conference on Language Resources and Evaluation (LREC) will be organised in 2010 by ELRA in cooperation with a wide range of international associations and organisations.

CONFERENCE AIMS

In 12 years – the first LREC was held in Granada in 1998 – LREC has become the major event on Language Resources (LRs) and Evaluation for Human Language Technologies (HLT). The aim of LREC is to provide an overview of the state-of-the-art, explore new R&D directions and emerging trends, exchange information regarding LRs and their applications, evaluation methodologies and tools, ongoing and planned activities, industrial uses and needs, requirements coming from the e-society, both with respect to policy issues and to technological and organisational ones.

LREC provides a unique forum for researchers, industrials and funding agencies from across a wide spectrum of areas to discuss problems and opportunities, find new synergies and promote initiatives for international cooperation, in support to investigations in language sciences, progress in language technologies and development of corresponding products, services and applications, and standards.

Special Highlight: Contribute to building the LREC2010 Map!

LREC2010 recognises that time is ripe to launch an important initiative, the LREC2010 Map of Language Resources, Technologies and Evaluation. The Map will be a collective enterprise of the LREC community, as a first step towards the creation of a very broad, community-built, Open Resource Infrastructure. As first in a series, it will become an essential instrument to monitor the field and to identify shifts in the production, use and evaluation of LRs and LTs over the years.

When submitting a paper, from the START page you will be asked to fill in a very simple template to provide essential information about resources (in a broad sense that includes technologies, standards, evaluation kits, etc.) that either have been used for the work described in the paper or are a new result of your research.

The Map will be disclosed at LREC, where some event(s) will be organised around this initiative.

CONFERENCE TOPICS

Issues in the design, construction and use of Language Resources (LRs): text, speech, other associated media and modalities

• Guidelines, standards, specifications, models and best practices for LRs

• Methodologies and tools for LRs construction and annotation

• Methodologies and tools for the extraction and acquisition of knowledge

• Ontologies and knowledge representation

• Terminology

• Integration between (multilingual) LRs, ontologies and Semantic Web technologies

• Metadata descriptions of LRs and metadata for semantic/content markup

• Validation, quality assurance, evaluation of LRs

Exploitation of LRs in different types of systems and applications

• For: information extraction, information retrieval, speech dictation, mobile communication, machine translation, summarisation, semantic search, text mining, inferencing, reasoning, etc.

• In different types of interfaces: (speech-based) dialogue systems, natural language and multimodal/multisensorial interactions, voice activated services, cognitive systems, etc.

• Communication with neighbouring fields of applications, e.g. e-government, e-culture, e-health, e-participation, mobile applications, etc.

• Industrial LRs requirements, user needs

Issues in Human Language Technologies evaluation

• HLT Evaluation methodologies, protocols and measures

• Benchmarking of systems and products

• Usability evaluation of HLT-based user interfaces (speech-based, text-based, multimodal-based, etc.), interactions and dialogue systems

• Usability and user satisfaction evaluation

General issues regarding LRs & Evaluation

• National and international activities and projects

• Priorities, perspectives, strategies in national and international policies for LRs

• Open architectures

• Organisational, economical and legal issues

PROGRAMME

The Scientific Programme will include invited talks, oral presentations, poster and demo presentations, and panels.

There is no difference in quality between oral and poster presentations. Only the appropriateness of the type of communication (more or less interactive) to the content of the paper will be considered.

SUBMISSIONS AND DATES

Submitted abstracts of papers for oral and poster or demo presentations should consist of about 1500-2000 words.

• Submission of proposals for oral and poster/demo papers: 31 October 2009

Proposals for panels, workshops and tutorials will be reviewed by the Programme Committee.

• Submission of proposals for panels, workshops and tutorials: 31 October 2009

PROCEEDINGS

The Proceedings on CD will include both oral and poster papers, in the same format. They will be added to the ELRA web archives before the conference.

A Book of Abstracts will be printed.

CONFERENCE PROGRAMME COMMITTEE

Nicoletta Calzolari, Istituto di Linguistica Computazionale del CNR - Pisa, Italy (Conference chair)

Khalid Choukri - ELRA, Paris, France

Bente Maegaard - CST, University of Copenhagen, Denmark

Joseph Mariani - LIMSI-CNRS and IMMI, Orsay, France

Jan Odijk - UIL-OTS, Utrecht, The Netherlands

Stelios Piperidis - Institute for Language and Speech Processing (ILSP), Athens, Greece

Mike Rosner – Department of Intelligent Computer Systems, University of Malta, Malta

Daniel Tapias - Sigma Technologies S.L., Madrid, Spain

4-6 . (2010-05-25) CfP JEP 2010

JEP 2010

XXVIIIèmes Journées d'Étude sur la Parole

Université de Mons, Belgique

du 25 au 28 mai 2010

http://w3.umh.ac.be/jep2010

=====================================================================

Les Journées d'Études de la Parole (JEP) sont consacrées à l'étude de la communication parlée ainsi qu'à ses applications. Ces journées ont pour but de rassembler l'ensemble des communautés scientifiques francophones travaillant dans le domaine. La conférence se veut aussi un lieu d'échange convivial entre doctorants et chercheurs confirmés.

En 2010, les JEP sont organisées par le Laboratoire des Sciences de la Parole de l'Académie Wallonie-Bruxelles, sur le site de l'Université de Mons en Belgique, sous l'égide de l'AFCP

(Association Francophone de la Communication Parlée) avec le soutien de l'ISCA (International Speech Communication Association).

Un second appel à communication précisant les thèmes ainsi que les modalités de soumission suivra ce premier appel.

CALENDRIER

===========

Date limite de soumission: 11 janvier 2010

Notification aux auteurs: 15 mars 2010

Conférence: 25-28 mai 2010

V. Delvaux

Chargée de Recherches FNRS

Laboratoire de Phonétique

Service de Métrologie et Sciences du Langage

Université de Mons-Hainaut

18, Place du Parc

7000 Mons

Belgium

+3265373140

http://staff.umh.ac.be/Delvaux.Veronique/index.html

5 . Books,databases and softwares

5-1 . Books

This section shows recent books whose titles been have communicated by the authors or editors.

Also some advertisements for recent books in speech are included.

This book presentation is written by the authors and not by this newsletter editor or any volunteer reviewer.

5-1-1 . Advances in Digital Speech Transmission

Advances in Digital Speech Transmission

Editors: Rainer Martin, Ulrich Heute and Christiane Antweiler

Publisher: Wiley&Sons

Year: 2008

5-1-2 . Sprachverarbeitung -- Grundlagen und Methoden der Sprachsynthese und Spracherkennung

Title: Sprachverarbeitung -- Grundlagen und Methoden

der Sprachsynthese und Spracherkennung

Authors: Beat Pfister, Tobias Kaufmann

Publisher: Springer

Year: 2008

Website: http://www.springer.com/978-3-540-75909-6

5-1-3 . Digital Speech Transmission

Digital Speech Transmission

Authors: Peter Vary and Rainer Martin

Publisher: Wiley&Sons

Year: 2006

5-1-4 . Distant Speech Recognition,

Distant Speech Recognition, Matthias Wölfel and John McDonough (2009), J. Wiley & Sons.

Please link the title to http://www.distant-speech-recognition.com

In the very recent past, automatic speech recognition (ASR) systems have attained acceptable performance when used with speech captured with a head-mounted or close-talking microphone (CTM). The performance of conventional ASR systems, however, degrades dramatically as soon as the microphone is moved away from the mouth of the speaker. This degradation is due to a broad variety of effects that are not found in CTM speech, including background noise, overlapping speech from other speakers, and reverberation. While conventional ASR systems underperform for speech captured with far-field sensors, there are a number of techniques developed in other areas of signal processing that can mitigate the deleterious effects of noise and reverberation, as well as separating speech from overlapping speakers. Distant Speech Recognition presents a contemporary and comprehensive description of both theoretic abstraction and practical issues inherent in the distant ASR problem.

5-1-5 . Automatic Speech and Speaker Recognition: Large Margin and Kernel Methods

Automatic Speech and Speaker Recognition: Large Margin and Kernel Methods

Joseph Keshet and Samy Bengio, Editors

John Wiley & Sons

March, 2009

Website: Automatic Speech and Speaker Recognition: Large Margin and Kernel Methods

About the book:

This is the first book dedicated to uniting research related to speech and speaker recognition based on the recent advances in large margin and kernel methods. The first part of the book presents theoretical and practical foundations of large margin and kernel methods, from support vector machines to large margin methods for structured learning. The second part of the book is dedicated to acoustic modeling of continuous speech recognizers, where the grounds for practical large margin sequence learning are set. The third part introduces large margin methods for discriminative language modeling. The last part of the book is dedicated to the application of keyword-spotting, speaker

verification and spectral clustering.

Contributors: Yasemin Altun, Francis Bach, Samy Bengio, Dan Chazan, Koby Crammer, Mark Gales, Yves Grandvalet, David Grangier, Michael I. Jordan, Joseph Keshet, Johnny Mariéthoz, Lawrence Saul, Brian Roark, Fei Sha, Shai Shalev-Shwartz, Yoram Singer, and Nathan Srebo.

5-1-6 . Some aspects of Speech and the Brain.

Some aspects of Speech and the Brain.

Susanne Fuchs, Hélène Loevenbruck, Daniel Pape, Pascal Perrier

Editions Peter Lang, janvier 2009

What happens in the brain when humans are producing speech or when they are listening to it ? This is the main focus of the book, which includes a collection of 13 articles, written by researchers at some of the foremost European laboratories in the fields of linguistics, phonetics, psychology, cognitive sciences and neurosciences.

5-1-7 . Spoken Language Processing,

Spoken Language Processing, edited by Joseph Mariani (IMMI and
LIMSI-CNRS, France). ISBN: 9781848210318. January 2009. Hardback 504 pp

Publisher ISTE-Wiley

Speech processing addresses various scientific and technological areas. It includes speech analysis and variable rate coding, in order to store or transmit speech. It also covers speech synthesis, especially from text, speech recognition, including speaker and language identification, and spoken language understanding. This book covers the following topics: how to realize speech production and perception systems, how to synthesize and understand speech using state-of-the-art methods in signal processing, pattern recognition, stochastic modeling, computational linguistics and human factor studies.

More on its content can be found at
http://www.iste.co.uk/index.php?f=a&ACTION=View&id=150

5-2 . Database providers

5-2-1 . LDC News

In this newsletter:

LDC2009T27

- Chinese Gigaword Fourth Edition -

LDC2009S03

- CSLU: S4X Release 1.2 -

LDC2009T23

- FactBank 1.0 -

- LDC’s Free Resources -

- Release of XTrans -

New Publications

(1) Chinese Gigaword Fourth Edition is a comprehensive archive of newswire text data that has been acquired over several years by the LDC. This edition includes all of the contents in Chinese Gigaword Third Edition (LDC2007T38) as well as newly collected data. In addition, four entirely new sources have been added in the fourth edition, Central News Service, Guangming Daily, People's Liberation Army Daily, and People's Daily.

The eight distinct international sources of Chinese newswire included in this edition are the following:

Agence France Presse
Central News Agency, Taiwan
Central News Service
Guangming Daily
People's Daily
People's Liberation Army Daily
Xinhua News Agency
Zaobao Newspaper

The original data received by the LDC from AFP, People's Liberation Army Daily, Xinhua, and Zaobao were encoded in GB-2312, those from CNA were in Big-5, and those from GMW, CNS, and People's Daily were in a combination of GB-2312 and GB-18030. To avoid the problems and confusion that could result from differences in character-set specifications, all text files in this corpus have been converted to UTF-8 character encoding.

New in the Fourth Edition:

Two years worth of new articles (January 2007 through December 2008) have been added to the Xinhua, Agence France Presse, and CNA data sets.
Four new data sources have been added - Guangming Daily, Central News Service , People's Daily, and People's Liberation Army daily, covering a timespan from November 2006 through December 2008.

Chinese Gigaword Fourth Edition is distributed on 1 DVD-ROM.

2009 Subscription Members will automatically receive two copies of this corpus. 2009 Standard Members may request a copy as part of their 16 free membership corpora. Non-members may license this data for US$5000.

(2) CSLU: S4X Release 1.2 was created by the Center for Spoken Language Understanding, Oregon Health and Science University (CSLU). The corpus consists of 36 speakers (22 male, 14 female) uttering 11 specified words. The speakers repeated the following words six times on each of four channels: startrek, supernova, tektronix, generation, nebula, processing, singularity, 71523, abracadabra, sungeeta and computer. The four channels used were office phone, home phone, carbon microphone telephone and speaker phone. Each speech file has a corresponding time-aligned phoneme-level transcription (achieved using automatic forced alignment) and an automatically-generated world-level transcription. Humans reviewed each utterance in two passes and classified it as good, bad, noisy or different.

The data was recorded with the CSLU T1 digital data collection system. Each utterance is recorded as a separate file. These files were sampled at 8 khz 8-bit and stored as ulaw files. All of the data use the RIFF standard file format. This file format is 16-bit linearly encoded.

CSLU: S4X Release 1.2 is distributed on one CD-ROM.

2009 Subscription Members will automatically receive two copies of this corpus, provided that they have submitted a signed copy of the LDC User Agreement for CSLU Corpora. 2009 Standard Members may request a copy as part of their 16 free membership corpora. Non-members may license this data for US$150.

(3) FactBank 1.0 consists of 208 documents (over 77,000 tokens) from newswire and broadcast news reports in which event mentions are annotated with their degree of factuality, that is, the degree to which they correspond to those events. FactBank 1.0 was built on top of TimeBank 1.2 and a fragment of the AQUAINT TimeML Corpus, both of which used the TimeML specification language. This resulted in a double-layered annotation of event factuality. TimeBank 1.2 and AQUAINT TimeML encode most of the basic structural elements expressing factuality information while FactBank 1.0 represents the resulting factuality interpretation. The combination of the factuality values in FactBank with the structural information in TimeML-annotated corpora facilitates the development of tools aimed at automatically identifying the factuality values of events, a component fundamental in tasks requiring some degree of text understanding, such as Textual Entailment, Question Answering, or Narrative Understanding.

FactBank annotations indicate whether the event mention describes actual situations in the world, situations that have not happened, or situations of uncertain interpretation. Event factuality is not an inherent feature of events but a matter of perspective. Different discourse participants may present divergent views about the factuality of the very same event. Consequently, in FactBank, the factuality degree of events is assigned relative to the relevant sources at play. In this way, it can adequately reflect the divergence of opinions regarding the factual status of events, as is common in news reports.

All FactBank markup is standoff and is represented through a set of 20 tables which can be easily loaded into a database. Each table resides in an independent text file, where fields are separated by three consecutive bars (i.e., |||). The data in fields of string type are presented between simple quotations ('). Because FactBank 1.0 was built on top of TimeBank 1.2 and AQUAINT TimeML, both of which are marked up with inline XML-based annotation, this release contains the TimeBank 1.2 and AQUAINT TimeML annotation in standoff, table-based format as well.

FactBank 1.0 is distributed via web download.

2009 Subscription Members will automatically receive two copies of this collection on disc. 2009 Standard Members may request a copy as part of their 16 free membership corpora. Non-members may license this data by completing the LDC User Agreement for Non-members. The agreement can be faxed to +1 215 573 2175 or scanned and emailed to this address. The collection is being made available at no charge.

LDC’s Free Resources

LDC is pleased to distribute FactBank 1.0 which is available at no cost. To license a copy of this data, non-members should complete the LDC User Agreement for Non-members and fax to +1 215 573 2175 or scan and email to this address. FactBank joins a host of LDC resources which are available for free. These resources include tools and corpora developed at LDC as well as corpora made available through LDC's strong network of data providers.

Since LDC's founding, we have distributed over 1300 copies of corpora at no cost including:

over 700 non-member downloads of Buckwalter Arabic Morphological Anaylzer 1.0
400 copies of Talkbank-sponsored data including popular releases such as the American National Corpus and the Santa Barbara Corpora of Spoken American English
nearly 200 copies of Web 1T 5-gram Version 1, sponsored by Google Inc.
over 30 copies of TimeBank 1.2
over a dozen copies of the corpora developed for the Unified Linguistic Annotation (ULA) project

For further information, visit our What's New! What's Free! Archive.

Release of XTrans

At InterSpeech 2009, LDC introduced XTrans, a new tool for manual transcription and annotation of audio recordings. XTrans is a next generation transcription tool that is designed to support transcription tasks in multiple languages on multiple platforms. XTrans provides a flexible and intuitive graphical user interface for a multitude of speech annotation tasks including (virtual) segmentation of audio into smaller units like turns and sentences; speaker identification; orthographic transcription in any language; and labeling of structural elements of the transcript like topics. Its versatile and powerful waveform display/playback component can load multiple audio files of different file formats and sampling rates at the same time. LDC and its partners have used XTrans to generate over 3500 hours of time-aligned verbatim transcripts in a variety of genres and languages.

With an intuitive interface, user configurability and embedded QC functions, XTrans is optimized for high-quality, high-volume transcription tasks involving real world data. XTrans successfully addresses the challenges of real world data including transcribing multiple speakers in a single channel through Virtual Speaker Channel, which enables an unlimited number of distinct speakers to be associated with the same audio channel. Furthermore, XTrans allows transcribers to open an effectively unlimited number of audio files for simultaneous transcription. Transcribers can switch focus between one, two or multiple speakers as needed. XTrans also provides strong multilingual support, with bidirectional text input for languages like Arabic, Farsi, Urdu, and Hebrew.

Realtime transcription rates have improved dramatically in LDC projects using XTrans, with rates for some tasks cut by as much as half. XTrans also brings key quality control functions directly into the interface, giving transcribers the power to improve the quality of their own work. XTrans components are written in Python and C++, utilizing LDC's QWave waveform display module. Even with very large files or multiple recordings, XTrans provides users with fast display and playback capabilities. A range of audio formats is supported, including .sph, .wav, .aiff, .flac, and .ogg. Transcripts are output in a Tab Delimited Format (TDF), which is easily converted to other common formats and is readily usable by downstream manual and automatic annotation tasks.

Availability:

XTrans for Linux and Windows platforms is available from the LDC at no cost under GPLv3 and can be downloaded here.

5-2-2 . ELDA/ELRA press release

   Press Release - Immediate
Paris, France, September, 3rd 2009

Distribution Agreement signed for BioLexicon

ELRA together with the European Bioinformatics Institute (EBI, Hinxton, UK), Istituto di Linguistica Computazionale-Consiglio Nazionale Ricerche (ILC-CNR, Pisa, Italy), and the National Centre for Text Mining (NaCTeM, University of Manchester, UK) has signed a Language Resources distribution agreement for a large-scale English language terminological resource in the biomedical domain: BioLexicon.

Biological terminology is a frequent cause of analysis errors when processing literature written in the biology domain, due largely to the high degree of variation in term forms, to the frequent mis-matches between labels of controlled vocabularies and ontologies on the one hand and the forms actually occurring in text on the other, and to the lack of detailed formal information on the linguistic behaviour of domain terms. For example, "retro-regulate" is a terminological verb often used in molecular biology but it is not included in conventional dictionaries. BioLexicon is a linguistic resource for the biology domain, tailored to cope with these problems. It contains information on:
    - terminological nouns, including nominalised verbs and proper names (e.g., gene names)
    - terminological adjectives
    - terminological adverbs
    - terminological verbs
    - general English words frequently used in the biology domain

Existing information on terms was integrated, augmented, complemented and linked, through processing of massive amounts of biomedical text, to yield inter alia over 2.2M entries, and information on over 1.8M variants and on over 2M synonymy relations. Moreover, extensive information is provided on how verbs and nominalised verbs in the domain behave at both syntactic and semantic levels, supporting thus applications aiming at discovery of relations and events involving biological entities in text.

This comprehensive coverage of biological terms makes BioLexicon a unique linguistic resource within the domain. It is primarily intended to support text mining and information retrieval in the biomedical
domain, however its standards-based structure and rich content make it a valuable resource for many other kinds of application.

On behalf of ELRA, ELDA will act as the distribution agency, by incorporating the BioLexicon in the ELRA Language Resources catalogue.

With these resources, ELRA is willing to extend the current catalogue, by offering specialized resources and thus allow a better coverage of the language.

For more information on BioLexicon (catalogue reference: ELRA-S0373): http://catalog.elra.info/product_info.php?products_id=1113

For more information on the ELRA catalogue, please contact:
Valérie Mapelli, mapelli@elda.org

For more information on ELRA & ELDA, please contact:
Khalid Choukri, choukri@elda.org
Hélène Mazo, mazo@elda.org

ELDA
55-57, rue Brillat Savarin
75013 Paris (France)

Tel.: +33 1 43 13 33 33
Fax: +33 1 43 13 33 30

5-2-3 . MEDAR project/ELDA

The goal of the MEDAR project, supported by the European Commission ICT programme, is to establish a network of partner centres of best practice in Arabic dedicated to promoting Arabic HLT (Human Language Technologies).

Within this framework, we are working on a global directory of players, experts, projects and Language resources related to Arabic Human Language Technology. If you have not already answered the 1st survey and would like to be part of the community, please take a few minutes to complete the survey: http://survey.elda.org/index.php?sid=15471&lang=en.
The collected information will be made available through a Knowledge Base.

About MEDAR: http://www.medar.info

5-2-4 . ELRA - Language Resources Catalogue - Update

*****************************************************************
ELRA - Language Resources Catalogue - Update
*****************************************************************

ELRA is happy to announce that 1 new Terminology Database, 3 Speech Desktop/Microphone resources and 1 Speech Telephone resource are now available in its catalogue:
*
**ELRA-T0373 BioLexicon *
BioLexicon is a large-scale English terminological resource which has been developed to address the needs emerging in text mining efforts in the biomedical domain. It contains over 2.2M lexical entries (over 3.3M semantic relations), and information on over 1.8M variants and on over 2M synonymy relations. BioLexicon is available in a relational database format (MySQL dump format) and it adheres to the EAGLES/ISO standards for lexical resources.
For more information, see: http://catalog.elra.info/product_info.php?products_id=1113

*ELRA-S0301 Norwegian EUROM1 (EUROM1_N)*
EUROM1 is the first really multilingual speech database produced in Europe. Over 60 speakers per language pronounced numbers, sentences, isolated words using close talking microphone.
For more information, see: http://catalog.elra.info/product_info.php?products_id=1114

*ELRA-S0302 TC-STAR female baseline voice: Laura*
Laura contains the recordings of one female English (British) speaker recorded in a noise-reduced room through a headset microphone. It consists of the recordings and annotations of read text material of approximately 10 hours of speech for baseline applications (Text-to-Speech systems).
For more information, see: http://catalog.elra.info/product_info.php?products_id=1115

*ELRA-S0303 TC-STAR male baseline voice: Ian*
Ian contains the recordings of one male English (British) speaker recorded in a noise-reduced room through a headset microphone. It consists of the recordings and annotations of read text material of approximately 10 hours of speech for baseline applications (Text-to-Speech systems).
For more information, see: http://catalog.elra.info/product_info.php?products_id=1116

*ELRA-S0304 SpeechDat(M) Italian Mobile Network Speech Database*
This speech database contains the recordings of 342 Italian speakers recorded over the Italian mobile telephone network. Each speaker uttered around 40 read and spontaneous items.
For more information, see: http://catalog.elra.info/product_info.php?products_id=1117

For more information on the catalogue, please contact Valérie Mapelli mailto:mapelli@elda.org

Visit our On-line Catalogue: http://catalog.elra.info
Visit the Universal Catalogue: http://universal.elra.info
Archives of ELRA Language Resources Catalogue Updates: http://www.elra.info/LRs-Announcements.html

5-2-5 . The ELRA Catalogue of Language Resources at OLAC

The ELRA Catalogue of Language Resources at OLAC

In the framework of its ongoing collaborative work with the OLAC community and in order to improve its compliance with OLAC standards, ELRA is pleased to announce that the OLAC archive of its Catalogue of Language Resources has been updated. According to OLAC standards, the ELRA Catalogue archive achieves the maximum rate of 5 stars on the OLAC scale.

The ELRA Catalogue has been exported into OLAC standards by means of associating the ELRA metadata for describing Language Resources with the OLAC metadata. Daily updates are performed through the automatic harvesting of the ELRA Catalogue which is carried out by the OLAC server.

With this work, ELRA aims at rendering its catalogue more visible to the wider community and it contributes to sharing information on Language Resources through standardized metadata sets.

To visit the ELRA Catalogue: http://catalogue.elra.info

and the corresponding OLAC archive: http://www.language-archives.org/archive/catalogue.elra.info

*** About ELRA ***
The European Language Resources Association (ELRA) is a non-profit making organisation founded by the European Commission in 1995, with the mission of providing a clearing house for language resources and promoting Human Language Technologies (HLT).

To find out more about ELRA, please visit our web site: http://www.elra.info

*** About OLAC ***
OLAC, the Open Language Archives Community, is an international partnership of institutions and individuals who are creating a worldwide virtual library of language resources by: (i) developing consensus on best current practice for the digital archiving of language resources, and (ii) developing a network of interoperating repositories and services for housing and accessing such resources.

To find out more about OLAC, please visit the following website: http://www.language-archives.org. From that website, ELRA resources can be searched, alongside resources from 40 other language archives.

6 . Jobs openings

We invite all laboratories and industrial companies which have job offers to send them to the ISCApad editor: they will appear in the newsletter and on our website for free. (also have a look at http://www.isca-speech.org/jobs.html as well as http://www.elsnet.org/ Jobs).

The ads will be automatically removed from ISCApad after 6 months. Informing ISCApad editor when the positions are filled will avoid irrelevant mails between applicants and proposers.

6-1 . (2009-05-04) Several Ph.D. positions and Ph.D. or Postdoc scholarships, Universität Bielefeld

Several Ph.D. Positions and Ph.D. or Postdoc Scholarships, Universität Bielefeld

Applications are invited for several Ph.D. positions and Ph.D. scholarships in experimental phonetics, speech technology and laboratory phonology at Universität Bielefeld (Fakultät für Linguistik und Literaturwissenschaft), Germany.

Successful candidates should hold a Master's degree (or equivalent) in phonetics, computational linguistics, linguistics, computer science or a related discipline. They will have a strong background in either

- speech synthesis and/or recognition

- discourse prosody

- laboratory phonology

- speech and language rhythm research

- multimodal speech (technology)

Candidates should appreciate working in an interdisciplinary environment. Good knowledge in experimental design techniques and programming skills will be considered a plus. Strong interest in research and high proficiency in English is required.

The Ph.D. positions will be part-time (50%); salary and social benefits are determined by the German public service pay scale (TVL-E13). The Ph.D. scholarship is based on the DFG scale. There is no mandatory teaching load.

Bielefeld University is an equal opportunity employer. Women are therefore particularly encouraged to apply. Disabled applicants with equivalent qualification will be treated preferentially.

The positions are available for three years (with a potential extension for the Ph.D. positions), starting as soon as

possible. Please submit your documents (cover letter, CV including list of publications, statement of research interests, names of two referees) electronically to the address indicated below. Applications must be received by June 15, 2009.

Universität Bielefeld

Fakultät für Linguistik und Literaturwissenschaft

Prof. Dr. Petra Wagner

Postfach 10 01 31

33 501 Bielefeld

Germany

Email: petra.wagner@uni-bielefeld.de

6-2 . (2009-05-07) PhD POSITION in MACHINE TRANSLATION AND SPEECH UNDERSTANDING FRANCE)

=============================================================================
PhD POSITION in MACHINE TRANSLATION AND SPEECH UNDERSTANDING (starting
09/09)
=============================================================================

The PORT-MEDIA (ANR CONTINT 2008-2011) is a cooperative project
sponsored by the French National Research Agency, between the University
of Avignon, the University of Grenoble, the University of Le Mans, CNRS
at Nancy and ELRA (European Language Resources Association). PORT-MEDIA
will address the multi-domain and multi-lingual robustness and
portability of spoken language understanding systems. More specifically,
the overall objectives of the project can be summarized as:
- robustness: integration/coupling of the automatic speech recognition
component in the spoken language understanding process.
- portability across domains and languages: evaluation of the genericity
and adaptability of the approaches implemented in the
understanding systems, and development of new techniques inspired by
machine translation approaches.
- representation: evaluation of new rich structures for high-level
semantic knowledge representation.

The PhD thesis will focus on the multilingual portability of speech
understanding systems. For example, the candidate will investigate
techniques to fast adapt an understanding system from one language to
another and creating low-cost resources with (semi) automatic methods,
for instance by using automatic alignment techniques and lightly
supervised translations. The main contribution will be to fill the gap
between the techniques currently used in the statistical machine
translation and spoken language understanding fields.

The thesis will be co-supervised by Fabrice Lefèvre, Assistant Professor
at LIA (University of Avignon) and Laurent Besacier, Assistant Professor
at LIG (University of Grenoble). The candidate will spend 18 months at
LIG then 18 months at LIA.

The salary of a PhD position is roughly 1,300€ net per month. Applicants
should hold a strong university degree entitling them to start a
doctorate (Masters/diploma or equivalent) in a relevant discipline
(Computer Science, Human Language Technology, Machine Learning, etc).
The applicants should be fluent in English. Competence in French is
optional, though applicants will be encouraged to acquire this skill
during training. All applicants should have very good programming skills.

For further information, please contact Fabrice Lefèvre (Fabrice.Lefevre
at univ-avignon.fr) AND Laurent Besacier (Laurent.Besacier at imag.fr).

====================================================================================
Sujet de thèse en Traduction Automatique et Compréhension de la Parole
(début 09/09)
====================================================================================

Le projet PORT-MEDIA (ANR CONTINT 2008-2011) concerne la robustesse et
la portabilité multidomaine et multilingue des systèmes de compréhension
de l'oral. Les partenaires sont le LIG, le LIA, le LORIA, le LIUM et
ELRA (European Language Ressources Association). Plus précisément, les
trois objectifs principaux du projet concernent :
-la robustesse et l'intégration/couplage du composant de reconnaissance
automatique de la parole dans le processus de compréhension.
-la portabilité vers un nouveau domaine ou langage : évaluation des
niveaux de généricité et d'adaptabilité des approches implémentées dans
les systèmes de compréhension.
-l’utilisation de représentations sémantiques de haut niveau pour
l’interaction langagière.

Ce sujet de thèse concerne essentiellement la portabilité multilingue
des différents composants d’un système de compréhension automatique ;
l’idée étant d’utiliser, par exemple, des techniques d’alignement
automatique et de traduction pour adapter rapidement un système de
compréhension d’une langue vers une autre, en créant des ressources à
faible coût de façon automatique ou semi-automatique. L'idée forte est
de rapprocher les techniques de traduction automatique et de
compréhension de la parole.

Cette thèse est un co-encadrement entre deux laboratoires (Fabrice
Lefevre, LIA & Laurent Besacier, LIG). Les 18 premiers mois auront lieu
au LIG, les 18 suivants au LIA.

Le salaire pour un etudiant en thèse est d'environ 1300€ net par mois.
Nous recherchons des étudiants ayant un Master (ou équivalent) mention
Recherche dans le domaine de l'Informatique, et des compétences dans les
domaines suivants : traitement des langues écrites et/ou parlées,
apprentissage automatique...

Pour de plus amples informations ou candidater, merci de contacter
Fabrice Lefèvre (Fabrice.Lefevre at univ-avignon.fr) ET Laurent Besacier
(Laurent.Besacier at imag.fr).

--------------------------

6-3 . (2009-05-07)Several Ph.D. Positions and Ph.D. or Postdoc Scholarships, Universität Bielefeld

Several Ph.D. Positions and Ph.D. or Postdoc Scholarships, Universität Bielefeld

- speech synthesis and/or recognition

- discourse prosody

- laboratory phonology

- speech and language rhythm research

- multimodal speech (technology)

Bielefeld University is an equal opportunity employer. Women are therefore particularly encouraged to apply. Disabled applicants with equivalent qualification will be treated preferentially.

The positions are available for three years (with a potential extension for the Ph.D. positions), starting as soon as

Universität Bielefeld

Fakultät für Linguistik und Literaturwissenschaft

Prof. Dr. Petra Wagner

Postfach 10 01 31

33 501 Bielefeld

Germany

Email: petra.wagner@uni-bielefeld.de

6-4 . (2009-05-08)PhD POSITION in MACHINE TRANSLATION AND SPEECH UNDERSTANDING (starting 09/09)

PhD POSITION in MACHINE TRANSLATION AND SPEECH UNDERSTANDING (starting 09/09)
=============================================================================

The PORT-MEDIA (ANR CONTINT 2008-2011) is a cooperative project sponsored by the French National Research Agency, between the University of Avignon, the University of Grenoble, the University of Le Mans, CNRS at Nancy and ELRA (European Language Resources Association). PORT-MEDIA will address the multi-domain and multi-lingual robustness and portability of spoken language understanding systems. More specifically, the overall objectives of the project can be summarized as:
- robustness: integration/coupling of the automatic speech recognition component in the spoken language understanding process.
- portability across domains and languages: evaluation of the genericity and adaptability of the approaches implemented in the
understanding systems, and development of new techniques inspired by machine translation approaches.
- representation: evaluation of new rich structures for high-level semantic knowledge representation.

The PhD thesis will focus on the multilingual portability of speech understanding systems. For example, the candidate will investigate techniques to fast adapt an understanding system from one language to another and creating low-cost resources with (semi) automatic methods, for instance by using automatic alignment techniques and lightly supervised translations. The main contribution will be to fill the gap between the techniques currently used in the statistical machine translation and spoken language understanding fields.

The thesis will be co-supervised by Fabrice Lefèvre, Assistant Professor at LIA (University of Avignon) and Laurent Besacier, Assistant Professor at LIG (University of Grenoble). The candidate will spend 18 months at LIG then 18 months at LIA.

The salary of a PhD position is roughly 1,300€ net per month. Applicants should hold a strong university degree entitling them to start a doctorate (Masters/diploma or equivalent) in a relevant discipline (Computer Science, Human Language Technology, Machine Learning, etc). The applicants should be fluent in English. Competence in French is optional, though applicants will be encouraged to acquire this skill during training. All applicants should have very good programming skills.

For further information, please contact Fabrice Lefèvre (Fabrice.Lefevre at univ-avignon.fr) AND Laurent Besacier (Laurent.Besacier at imag.fr).

6-5 . (2009-05-11) Thèse Cifre indexation de données multimédia Institut Eurecom

Thèse Cifre indexation de données multimédia

These

DeadLine: 01/11/2009

merialdo@eurecom.fr

http://bmgroup.eurecom.fr/

The Multimedia Communications Department of EURECOM, in partnership the travel service provider company AMADEUS, invites applications for a PhD position on multimedia indexing. The goal of the thesis is to study new techniques to organize large quantities of multimedia information, specifically images and videos, for improving services to travelers. This includes managing images and videos from providers as well as from users about places, locations, events, etc… The approach will be based on the most recent techniques in multimedia indexing, and will benefit from the strong research experience of EURECOM in this domain, joint to the industrial experience of AMADEUS.

We are looking for very good and motivated students, with a strong knowledge in image and video processing, statistical and probabilistic modeling, for the theoretical part, and a good C/C++ programming ability for the experimental part. English is required. The successful candidate will be employed by AMADEUS in Sophia Antipolis, and will strongly interact with the researchers at EURECOM.

Applicants should email a resume, letter of motivation, and all relevant information to.with

Prof. Bernard Merialdo

merialdo@eurecom.fr

The project will be conducted within AMADEUS (http://www.amadeus.com/), a world leader in provision of solutions to the travel industry to manage the distribution and selling of travel services. The company is the leading Global Distribution System (GDS) and the biggest processor of travel bookings in the world. Their main development center is located in Sophia Antipolis, France, and employs more than 1200 engineers. The research will be supervised by EURECOM (http://www.eurecom.fr), a graduate school and research center in communication systems, whose activity includes corporate, multimedia and mobile communications. EURECOM currently counts about 20 professors, 10 post-docs, 170 MS and 60 PhD students, and is involved in many European research projects and joint collaborations with industry. EURECOM is also located in Sophia-Antipolis, a major European technology park for telecommunications research and development in the French Riviera.

6-6 . (2009-05-11)Senior Research Fellowship in Speech Perception and Language Development,MARCS Auditory Laboratories

Ref 147/09 Senior Research Fellowship in Speech Perception and Language Development, MARCS Auditory Laboratories, Australia

MARCS Auditory Laboratories is a multi-disciplinary research centre involved in research in auditory perception and cognition, particularly in the fields of speech and language, music, sound and action, and hearing and auditory processes.

MARCS is seeking a Senior Research Fellow with a background in psychology/behavioural science and specialisation in some or all of the following: speech perception, speech science, experimental phonetics, infant and child perception studies, speech production studies (eg with OPTOTRAK), psychophysics/psychoacoustics, cross-language studies; and experience in sophisticated methods of data analysis.

As this position is likely to involve working with children, 'Prohibited Persons' are not permitted to apply. The successful applicant will be required to authorise a screening check.

5 Year Fixed Term Contract , Bankstown Campus

Remuneration Package: Academic Level C $107,853 to $123,724 p.a. (comprising Salary $91,266 to $104,831 p.a., 17% Superannuation, and Leave Loading)

Position Enquiries: Professor Denis Burnham, (02) 9772 6677 or email d.burnham@uws.edu.au

Closing Date: The closing date for this position has been extended until 30 June 2009.

6-7 . (2009-06-02)Proposition de sujet de thèse 2009 Analyse de scènes de parole Grenoble France

Proposition de sujet de thèse 2009
Ecole Doctorale EDISCE (http://www-sante.ujf-grenoble.fr/edisce/)
Financement ANR (http://www.icp.inpg.fr/~schwartz/Multistap/Multistap.html)

Analyse de scènes de parole : le problème du liage audio-visuo-moteur à la lumière de données comportementales et neurophysiologiques

Deux questions importantes traversent les recherches actuelles sur le traitement cognitif de la parole : la question de la multisensorialité (comment se combinent les informations auditives et visuelles dans le cerveau) et celle des interactions perceptuo-motrices.

Une question manquante est selon nous celle du « liage » (binding) : comment dans ces processus de traitement auditif ou audiovisuel, le cerveau parvient-il à « mettre ensemble » les informations pertinentes, à éliminer les « bruits », à construire les « flux de parole » pertinents avant la prise de décision ? Plus précisément, les objets élémentaires de la scène de parole sont les phonèmes, et des modules spécialisés auditifs, visuels, articulatoires contribuent au processus d'identification phonétique, mais il n'a pas été possible jusqu'à présent d'isoler leur contribution respective, ni la manière dont ces contributions sont fusionnées. Des expériences récentes permettent d'envisager le processus d'identification phonétique comme étant de nature non hériarchique, et essentiellement instancié par des opérations associatives. La thèse consistera à développer d’autres paradigmes expérimentaux originaux, mais aussi à mettre en place des expériences de neurophysiologie et neuroimagerie (EEG, IRMf) disponibles au laboratoire et dans son environnement Grenoblois, afin de déterminer la nature et le fonctionnement des processus de groupement audiovisuel des scènes de parole, en relation avec le mécanismes de production.

Cette thèse se réalisera dans le cadre d’un projet ANR « Multistap » (Multistabilité et groupement perceptif dans l’audition et dans la parole » http://www.icp.inpg.fr/~schwartz/Multistap/Multistap.html). Ce projet fournira à la fois le support de financement pour la bourse de thèse, et un environnement stimulant pour le développement des recherches, en partenariat avec des équipes de spécialistes d’audition et de vision, de Paris (DEC ENS), Lyon (LNSCC) et Toulouse (Cerco).

Responsables
Jean-Luc Schwartz (DR CNRS, HDR) : 04 76 57 47 12,
Frédéric Berthommier (CR CNRS) : 04 76 57 48 28
Jean-Luc.Schwartz, Frederic.Berthommier@gipsa-lab.grenoble-inp.fr

6-8 . (2009-06-10) PhD in ASR in Le Mans France

PhD position in Automatic Speech Recognition
=====================================

Starting in september-october 2009.

The ASH (Attelage de Systèmes Hétérogènes) project is a project funded by the ANR (French National Research Agency). Three French academic laboratories are involved: LIUM (University of Le Mans), LIA (University of Avignon) and IRISA (Rennes).

The main objective of the ASH project is to define and experiment an original methodological framework for the integration of heterogeneous automatic speech recognition systems. Integrating heterogeneous systems, and hence heterogeneous sources of knowledge, is a key issue in ASR but also in many other applicative fields concerned with knowledge integration and multimodality.

Clearly, the lack of a generic framework to integrate systems operating with different viewpoints, different types of knowledge and at different levels is a strong limitation that needs to be overcome: the definition of such a framework is the fundamental challenge of this work.

By defining a rigorous and generic framework to integrate systems, significant scientific progresses are expected in automatic speech recognition. Another objective of this project is to enable the efficient and reliable processing of large data streams by combining systems on the y.
At last, we expect to develop an on-the-fly ASR system as a real-time demonstrator of this new approach.

The thesis will be co-supervised by Paul Deléglise, Professeur at LIUM, Yannick Estève, Assistant Professor at LIUM and Georges Linarès, Assistant Professor at LIA. The candidate will work at Le Mans (LIUM), but will regularly spend a few days in Avignon (LIA)

Applicants should hold a strong university degree entitling them to start a doctorate (Masters/diploma or equivalent) in a relevant discipline (Computer Science, Human Language Technology, Machine Learning, etc).

The applicants for this PhD position should be fluent in English or in French. Competence in French is optional, though applicants will be encouraged to acquire this skill during training. This position is funded by the ANR.

Strong software skills are required, especially Unix/linux, C, Java, and a scripting language such as Perl or Python.

Contacts:
Yannick Estève: yannick.esteve@lium.univ-lemans.fr
Georges Linarès: georges.linares@univ-avignon.fr

6-9 . (2009-06-17)Two post-docs in the collaboration between CMU (USA) and University-Portugal program

Two post-doctoral positions in the framework of the Carnegie Mellon

University-Portugal program are available at the Spoken Language Systems

Lab (www.l2f.inesc-id.pt), INESC-ID, Lisbon, Portugal.

Positions are for a fixed term contract of length up to two and a half

years, renewable in one year intervals, in the scope of the research

projects PT-STAR (Speech Translation Advanced Research to and from

Portuguese) and REAP.PT (Computer Aided Language Learning – Reading

Practice), both financed by FCT (Portuguese Foundation for Science and

Technology).

The starting date for these positions is September 2009, or as soon as

possible thereafter.

Candidates should send their CVs (in .pdf format) before July 15th, to

the email addresses given below, together with a motivation letter.

Questions or other clarification requests should be emailed to the same

addresses.

======== PT-STAR (project CMU-PT/HuMach/0039/2008) ========

Topic: Speech-to-Speech Machine Translation

Description: We seek candidates with excellent knowledge in statistical

approaches to machine translation (and if possible also speech

technologies) and strong programming skills. Familiarity with the

Portuguese language is not at all mandatory, although the main source

and target languages are Portuguese/English.

Email address for applications: lcoheur at l2f dot inesc-id dot pt

======== REAP.PT (project CMU-PT/HuMach/0053/2008) ========

Topic: Computer Aided Language Learning

Description: We seek candidates with excellent knowledge in automatic

question generation (multiple-choice synonym questions, related word

questions, and cloze questions) and/or measuring the reading difficulty

of a text (exploring the combination of lexical features, grammatical

features and statistical models). Familiarity with a romance language is

recommended, since the target language is Portuguese.

Email address for applications: nuno dot mamede at inesc-id dot pt

6-10 . (2009-06-19) POSTDOC POSITION in SPEECH RECOGNITION FOR UNDER-RESOURCED LANGUAGES

POSTDOC POSITION in SPEECH RECOGNITION FOR UNDER-RESOURCED LANGUAGES (18 months ; starting January 2010 or later) IN GRENOBLE (France)

=============================================================================

PI (ANR BLANC 2009-2012) is a cooperative project sponsored by the French National Research Agency, between the University of Grenoble (France), the University of Avignon (France), and the International Research Center MICA in Hanoï (Vietnam).

PI addresses spoken language processing (notably speech recognition) for under-resourced languages (or ?-languages). From a scientific point of view, the interest and originality of this project consists in proposing viable innovative methods that go far beyond the simple retraining or adaptation of acoustic and linguistic models. From an operational point of view, this project aims at providing a free open source ASR development kit for ?-languages. We plan to distribute and evaluate such a development kit by deploying ASR systems for new under-resourced languages with very poor resources from Asia (Khmer, Lao) and Africa (Bantu languages).

The POSTDOC position focus on the development of ASR for two low-resourced languages from Asia and Africa. This includes : supervising the resource collection (in relation with the language partners), propose innovative methods to quickly develop ASR systems for these languages, evaluation., etc.

The salary of the POSTDOC position is roughly 2300€ net per month. Applicants should hold a PhD related to spoken language processing. The applicants should be fluent in English. Competence in French is optional, though applicants will be encouraged to acquire this skill during the postdoc.

For further information, please contact Laurent Besacier (Laurent.Besacier at imag.fr).

6-11 . (2009-06-22) PhD studentship in speech and machine learning ESPCI ParisTech

Thesis topic : Vocal Prosthesis Based on Machine Learning

The objective of the thesis is to design and implement a vocal prosthesis to restore the original voice of persons who have lost the ability to speak due to a partial or total laryngectomy or a neurological problem. Using a miniature ultrasound machine and a video camera to drive a speech synthesizer, the device is intended to restore the original voice of these patients with as much fidelity as possible, allowing speech handicapped individuals to interact with those around them in a more natural and familiar way. The thesis work will build upon promising results obtained in the Ouisper project (funded on contract number ANR-06-BLAN-0166, http://www.neurones.espci.fr/ouisper/index.htm, and also supported by the French Defense Department, DGA), which terminates at the end of 2009; however, final success will require addressing the following four key technological issues:

1) New data acquisition protocol: the current acquisition system requires the user’s head to be immobilized during speech. The candidate will need to design and implement an innovative new system to overcome this constraint, which would be unacceptable for a real world application.

2) New dictionaries: results obtained thus far show that a truly open domain vocabulary may not be realistic. The candidate will create new dictionaries of vocabularies which are constrained, yet rich enough to be of genuine utility for verbal communication in the targeted speech handicapped community.

3) New synthesis methods: concatenative synthesis, though conceptually simple, is not sufficiently flexible when the initial recognition step contains errors. The candidate will devise new synthesis methods which better model the spectral qualities of the speaker’s voice, perhaps using Bayesian networks. Innovative techniques of recovering an acceptable prosody for the synthesized speech will also need to be developed.

4) Real time execution: as the amount of calculation necessary to carry out the recognition and synthesis steps is significant, the candidate will need to pay particular attention to optimization of code and real time execution in the algorithms he or she develops.

The thesis will be carried out in partnership with the Laboratoire de Phonétique et de Phonologie of the Université de Paris III, specializing in speech production and pathologies, for which additional funding has been obtained from the Agence Nationale de la Recherche (Emergence-TEC 2009 call, REVOIX project).

6-12 . (2009-06-30)Postdoctoral Fellowships in machine learning/statistics/machine vision at Monash University, Australia

6-13 . (2009-06-30) PhD studentship at LIMSI France

Titre :

Modèles de l'expressivité pour la synthèse de récits courts, lus par un robot humanoïde.

Contenu:

Si les systèmes de synthèses actuels sont généralement suffisants pour lire des phrases de

façon neutre, il sont très vite pénibles à écouter, en particulier pour des textes assez long

(plusieurs paragraphes). Les systèmes de synthèse ne sont guère capables de rendre expressif

une narration. De même, les capacités motrices des robots humanoïdes sont actuellement peu

exploitées et développées pour l’expression par le geste et la posture.

Ce projet de recherche porte sur la synthèse expressive audiovisuelle de récits courts. Le

projet comprend deux aspects principaux. Dans une phase d’analyse, il s’agit de traiter

automatiquement des textes, courts récits de type « contes pour enfant », afin d'en extraire un

contenu pragmatique, sémantique, dialogique, narratif et émotionnel.

Ce contenu servira dans une seconde phase d'une part à la synthèse de prosodie expressive, et

d'autre part à alimenter un modèle comportemental en termes de postures, de gestes et autres

mouvements du robot humanoïde NAO.

Compétences requises:

Ce sujet est situé dans le domaine de l’interaction homme-machine expressive.

Il demande de posséder ou d'acquérir des compétences en informatique linguistique, tant du

point de vue de l'écrit que de l'oral, et si possible également du point de vue audio visuel.

Le projet contient une part significative de programmation pour l’analyse des texte et la

synthèse, mais aussi une part significative d’analyse linguistique (des textes), phonétique (de

la prosodie), et comportementale (posture et gestes).

Des profils de type informatique, science cognitive ou linguistique seront donc considérés.

Contexte et équipe d’accueil:

Cette thèse s’inscrit dans le contrat ANR GV-LEX.

Elle se déroulera au LIMSI-CNRS www.limsi.fr dans les groupes Audio & Acoustique,

Traitement du Langage Parlé, et Architecture et Modèles de l’Interaction

Cette thèse commencera dès septembre, financée par l’ANR pour une durée 3 ans.

Encadrement - contact:

La thèse sera encadrée par Christophe d’Alessandro, directeur de recherche au CNRS. Les

candidatures seront adressées aux quatre chercheurs impliqués dans ce projet :

Christophe d’Alessandro <cda@limsi.fr>

Jean-Claude Martin <martin@limsi.fr>

Sophie Rosset <sophie.rosset@limsi.fr>

Albert Rilliard <rilliard@limsi.fr>

6-14 . (2009-07-01) These: Vocal Prosthesis Based on Machine Learning (France)

  Vocal Prosthesis Based on Machine Learning(2)

These

DeadLine: 01/09/2009

denby@ieee.org

http://

We are looking for an excellent candidate for a PhD studentship in speech and statistical learning at the Laboratoire d'Electronique at ESPCI ParisTech, Paris, France. Interested candidates should contact Prof. B. Denby by mail at denby@ieee.org before 1 Septembre 2009 at the latest (earlier application is strongly encouraged).

Working language: French or English

Thesis topic : Vocal Prosthesis Based on Machine Learning

http://www.neurones.espci.fr/ouisper/index.htm,

and also supported by the French Defense Department, DGA), which terminates at the end of 2009; however, final success will require addressing the following four key technological issues:

1) New data acquisition protocol: the current acquisition system requires the users head to be immobilized during speech. The candidate will need to design and implement an innovative new system to overcome this constraint, which would be unacceptable for a real world application.

3) New synthesis methods: concatenative synthesis, though conceptually simple, is not sufficiently flexible when the initial recognition step contains errors. The candidate will devise new synthesis methods which better model the spectral qualities of the speaker's voice, perhaps using Bayesian networks. Innovative techniques of recovering an acceptable prosody for the synthesized speech will also need to be developed.

The thesis will be carried out in partnership with the Laboratoire de Phonetique et de Phonologie of the Universite de Paris III, specializing in speech production and pathologies, for which additional funding has been obtained from the Agence Nationale de la Recherche (Emergence-TEC 2009 call, REVOIX project).

http://gdr-isis.org/rilk/gdr/Kiosque/poste.php?jobid=3369

6-15 . (2009-07-06) PhD in SPEECH RECOGNITION FOR UNDER-RESOURCED LANGUAGES (Grenoble France)

POSTDOC POSITION in SPEECH RECOGNITION FOR UNDER-RESOURCED LANGUAGES (18 months ; starting January 2010 or later) IN GRENOBLE (France)
=============================================================================

PI (ANR BLANC 2009-2012) is a cooperative project sponsored by the French National Research Agency, between the University of Grenoble (France), the University of Avignon (France), and the International Research Center MICA in Hanoï (Vietnam).

PI addresses spoken language processing (notably speech recognition) for under-resourced languages (or ?-languages). From a scientific point of view, the interest and originality of this project consists in proposing viable innovative methods that go far beyond the simple retraining or adaptation of acoustic and linguistic models. From an operational point of view, this project aims at providing a free open source ASR development kit for ?-languages. We plan to distribute and evaluate such a development kit by deploying ASR systems for new under-resourced languages with very poor resources from Asia (Khmer, Lao) and Africa (Bantu languages).

The POSTDOC position focus on the development of ASR for two low-ressourced languages from Asia and Africa. This includes : supervising the ressource collection (in relation with the language partners), propose innovative methods to quickly develop ASR systems for these languages, evaluation., etc.

The salary of the POSTDOC position is roughly 2300? net per month. Applicants should hold a PhD related to spoken language processing. The applicants should be fluent in English. Competence in French is optional, though applicants will be encouraged to acquire this skill during the postdoc.

For further information, please contact Laurent Besacier (Laurent.Besacier at imag.fr).

6-16 . (2009-07-08) Position at Deutsche Telekom R&D

Deutsche Telekom, one of the world’s leading telecommunications and information

technology service provider, is expanding its corporate research and development

activities at Deutsche Telekom Inc., R&D Lab USA, Los Altos, California. Having a close

collaboration with top-notch institutions, the laboratories offer an unprecedented

combination of academic and industrial research with opportunities to have a direct

impact on company’s products and services.

There is a current opening for a highly qualified Senior Research Scientist the

research field New Media for the area of Multimedia Communications and Systems.

We are looking for a self-driven and motivated individual who is passionate about

conducting leading-edge research. Applicants should have recently completed a

doctoral degree in computer science, electrical engineering, or other related disciplines

and have expertise in different facets of multimedia communications such as media

coding, streaming, and compression, with hands on system building experience and

know-how of standards. Experience in industrial R&D will be valued.

Application material should include, in a single pdf file, the following in stated order, (a)

cover letter, (b) one-page statement of research objectives, (c) curriculum vitae, (d) list

of publications, and (e) contact information of at least three individuals who may serve

as references. Short-listed candidates will be invited to give a talk and have interviews

with members of the recruiting committee.

Please submit your application until 22 July 2009.

Deutsche Telekom Inc. is an equal opportunity employer.

Applications should be submitted via email to:

Dr. Jatinder Pal Singh

Deutsche Telekom Inc., R&D Lab USA

Email: laboratories.researchscientist@telekom.de

6-17 . (2009-07-15) PhD at LIMSI Paris