ISCApad number 106

April 10th, 2007

MESSAGE from David House, Secretary of the Board,

Dear members,
This month it is my turn to tell you something about myself and what I do on the ISCA board. My background is in phonetics, prosody and speech perception, and I am a professor of acoustic and phonetic analysis at KTH in Stockholm, Sweden. As ISCA secretary, my duties are predominately administrative in nature, a large part of which involves planning and keeping records of the activities and decisions of the ISCA board. We meet for two full days each year in conjunction with INTERSPEECH and have three or four teleconferences during the year. I take the minutes of our meetings, assemble an action list to help us all remember what we need to do, keep this list updated as the actions of the board are completed, and then receive input and suggestions to help compile the agenda for the next board meeting.
Another one of my task is running our elections. We will need to elect five new board members this July, and you will all soon receive more information by email about nominations and voting. I would like to encourage you all, as members of ISCA, to vote in the election for the new board members.
To help me in all these tasks, I have the excellent and very valuable support of Manu Foxonet who also runs the day-to-day operations of the Secretariat and with whom many of you have corresponded concerning your membership and other questions.
Finally I would like to add on a personal note that working on the Board of ISCA has given me a much deeper appreciation of the importance of ISCA for the field of international speech communication research, science and technology, and the value of our conferences and workshops. I hope to see you all soon at one of our workshops or in Antwerp in August for INTERSPEECH.

David House,
Professor at KTH, Stockholm
Editorial

Dear Members,
As nature, speech activity is awakening after winter. This spring issue contains a lot of job announcements and scholarships. Also new conferences,workshops and summer schools are under preparation. Pay attention to the submission deadlines: I do not always receive the information in time ( at least a month is necessary for a monthly newsletter!).
Some of you will attend ICASSP2007: I wish them a fruitful conference and ...an enjoyable trip!

Christian Wellekens

TABLE OF CONTENTS

  1. ISCA News
  2. SIG's activities
  3. Courses, internships
  4. Books, databases, softwares
  5. Job openings
  6. Journals
  7. Future Interspeech Conferences
  8. Future ISCA Tutorial and Research Workshops (ITRW)
  9. Forthcoming Events supported (but not organized) by ISCA
  10. Future Speech Science and technology events

ISCA NEWS


top

Call for applications to the Christian Benoit Award
The Christian Benoit Award is delivered periodically by the Association Christian Benoit (**). It is given to promising young scientists in the domain of Speech Communication. The Award provides financial support for the development of a multi-media project promoting the work of these young scientists, and is valued at 7,622 Euros.
The first award was delivered to Tony Ezzat from MIT in June 2000, for his research in Audiovisual Speech Synthesis, the second award to Johanna Barry from University of Melbourne in September 2002 for her work on the acquisition of lexical tones in profoundly hearing-impaired speakers using a cochlear implant, and the third award to Olov Engwal from KTH in Stockholm in October 2004 for the elaboration of ARTUR, a multi-modal articulation tutor able to give automatic feedback to real users.
The fourth award will be delivered this year to ANY PROJECT IN THE FIELD OF SPEECH COMMUNICATION. Candidates should be in the final stages of their doctoral research or within the five years following the obtention of their PhD.
The Christian Benoit award will offer financial support to develop a multi-media project which (a) demonstrates the candidate's research in a way that helps launching that candidate's career, and (b) leverages electronic publishing technologies intelligently so as to facilitate the widest possible dissemination of this content.
In the application, the candidate should provide
-- a statement of research interest,
-- a detailed curriculum vitae, and
-- a description of the proposed multi-media project.
If the project already exists, a copy or link should be provided along with the application.
Applications should be sent to Pascal Perrier and received by Friday April 20th, 2007. Electronic submissions are mandatory.
The successful candidate will be notified by June 1st and invited to make a brief presentation of his/her work at the Interspeech 2007 Conference in Antwerp (Belgium).
Travel expenses for attendance at the Award ceremony will be provided by the Christian Benoit Association. For further information, please contact Pascal Perrier.
** The Christian Benoit Association is a nonprofit organization, whose purpose is to facilitate the development of research projects in the field of speech communication. Established in honor of Christian Benoit, French CNRS researcher in the field of speech communication who died on the 26th of April, 1998, at the age of 41, the Award places special emphasis on multimedia representations of ongoing research.

Appel a participation au Prix de these AFCP
Association Francophone de la Communication Parlee
L'Association Francophone de la Communication Parlee (AFCP) decerne chaque annee un prix scientifique pour une excellente these du domaine. L'AFCP souhaite ainsi promouvoir toutes les facettes de la recherche en communication parlee : des travaux fondamentaux aux travaux appliques, du domaine des STIC, SHS ou SDV. L'objectif de ce prix est de dynamiser les jeunes chercheurs, tout en faisant connaitre leurs travaux.
Le jury est compose des chercheurs elus du CA de l'AFCP.
- R. Ridouane est laureat du prix 2004 pour: "Suite de consonnes en berbere: phonetique et phonologie",
- M. Dohen en 2005 pour: "Deixis prosodique multisensorielle: production et perception audiovisuelle de la focalisation contrastive en francais"
Details rubrique 'Bourses et Prix'.
La remise officielle du prix se fera lors de la rencontre francophone "Les Journees d'Etudes sur la Parole" (JEP) 2008 a Avignon. Chaque recipiendaire se verra remettre la somme de 500 euros, et sera invite a resumer ses travaux lors d'un expose.
CALENDRIER :
Peut candidater tout docteur ayant soutenu sa these entre le 1er octobre 2005 et le 31 decembre 2006. On ne peut candidater qu'a une seule edition.
Candidature avant le 20 avril 2007 (depot de la these & envoi postal).
Resultats : debut juin 2007.
DOSSIER DE CANDIDATURE :
1/ Deposez votre these (.pdf) sur le serveur AFCP des theses qui regroupe la plupart des theses francophones du domaine: www.afcp-parole.org/theses.htm
2/ Postez un CD a:
Herve' Glotin Prix AFCP UMR CNRS LSIS
Univ. Sud Toulon Var, BP20132
83957 La Garde Cedex 20 - France
contenant un seul fichier (votre nom.pdf) avec dans l'ordre:
* resume de votre these (2 pages),
* liste de vos publications,
* tous les rapports (jury et rapporteurs) scannes de votre soutenance de these,
* une lettre de recommandation scannee de votre directeur de these pour ce prix,
* votre CV (avec coord. completes dont Email).

SIG's activities


A list of Speech Interest Groups can be found on our web.

top

COURSES, INTERNSHIPS


Master in Human Language Technologies and Interfaces at the University of Trento

Website
organized by: University of Trento and Fondazione Bruno Kessler Irst
Call for applications, Academic Year 2007/08
Goal
Human language technology gives people the possibility of using speech and/or natural language to access a variety of automated services, such as airline reservation systems or voicemail, to access and communicate information across different languages, and to keep under control the increasing amount of information available by automatically extracting useful content and summarizing it. This master aims at providing skills in the basic theories, techniques, and applications of this technology through courses taught by internationally recognized researchers from the university, research centers and supporting industry partners. Students enrolled in the master will gain in depth knowledge from graduate courses and from substantial practical projects carried out in research and industry labs.
Courses:
Speech Processing, Machine Learning for NLP, Human Language, Text Processing, Spoken Dialog Systems,Human Computer Interaction, Language Resources, Multilingual Technology
Requisites
Master degree level ( min 4 years) in the area of computer science, electrical engineering, computational linguistics and cognitive science and other related disciplines. English language (official language)
Student Grants
A limited number of fellowships will be available.
Application Deadline
Non EU Students: June, 15
EU Students: end of July
Info
E-mail
University of Trento-Department of Information and Communication
Technologies Via Sommarive, 14-38100 Povo (Trento), Italy

Up to three PhD scholarships starting in October 2007 are available in the Partnership International for Research and Education (PIRE)

Meaning Representations in Language Understanding
The partnership for research and education (PIRE), established in 2005, is a collaborative PhD programme between
* Saarland University, Germany
* the Brown Laboratory for Linguistic Information Processing headed by Eugene Charniak
* The Johns Hopkins University Center for Language and Speech Processing (CLSP) headed by Frederick Jelinek
* Charles University (Jan Hajic), Czech Republic.
PIRE is also affiliated with our existing International Research Training Group (IRTG) co-operation with Edinburgh University.
Each scholarship is funded for two years, normally extendable for a third year. Doctoral degrees may be obtained in computational linguistics, phonetics, or informatics, from Saarland University. The official language of the programme is English. Distinguishing features of the cooperation include:
- Joint supervision of dissertations by lecturers from Saarbruecken and our US partner universities
- A six month research stay either at Brown University or at Johns Hopkins University
- An intensive research exchange programme between all four universities involved in PIRE.
PhD projects will be in the area of meaning representation for natural language processing and suitable applications like machine translation or speech reconstruction.
The scholarship currently provides EURO 1468 per month. Additional compensation includes family allowance (where applicable), travel funding, support for carrying out experiments, and an additional monthly allowance for the duration of the stay in the US. Applicants should hold a strong university degree equivalent to the German Diplom or Magister (e.g. Master's level) in a relevant discipline, and should not be more than 28 years of age. Women and international students are particularly encouraged to apply.
Deadline for applications: 30 April, 2007
Applications should be sent electronically
To be fully considered, please include:
- a curriculum vitae indicating degrees obtained, disciplines covered (e.g. list of courses ), publications, and other relevant experience
- a sample of written work (e.g. research paper, or dissertation, preferably in English)
- copies of high school and university certificates, and transcripts
- two references (e-mailed directly to the PIRE office (verburg@CoLi.Uni-SB.DE) before the deadline)
- a statement of research interests, previous knowledge and activities in any of the relevant research areas.
If you have already applied for our IRTG postions (http://www.coli.uni-saarland.de/projects/igk/announcement/) or are planing to do so, you do not need to send your complete application again, but just a short note, that you would also like to be considered for PIRE.
In case an application can only be submitted by regular post, it should be sent to:
PIRE office
Claudia Verburg
Department of Computational Linguistics
Saarland University
P.O. Box 15 11 50
D-66041 Saarbruecken
Germany
If you have any questions, please contact Prof. Dr. Dietrich Klakow For more information see also
http://www.pire.uni-saarland.de/index.html
http://www.clsp.jhu.edu/research/pire/
http://www.coli.uni-saarland.de/projects/igk/

Summer school: Cognitive and physical models of speech production, perception, and perception-production interaction.

Part II : Brain and Speech
Autrans, France
September 16-21, 2007
After the success of the previous summer school held in Lubmin (Germany) 2004, we are happy to announce the second international summer school on Cognitive and physical models of speech production, perception, and perception-production interaction. This year we will pay special attention to the brain. The aim of this summer school is to relate fundamental knowledge on speech production and perception to insights about the organization and function of the brain. Tutorials will be presented by specialists in these domains.
This summer school is intended mainly for graduate students, postdoctoral fellows, and researchers who work in the fields of speech production, perception, perception-production interaction, and the brain (neurolinguistics). Potential topics are:
Speech and language acquisition
Speech and language disorders
Neural basis of speech production
Speech production control
Neural basis of speech perception
Audio-visual speech perception
Plasticity of speech perception
It is intended to provide a platform for interchanges between students, junior and senior researchers, and hence, we would like each participant to feel free to contribute to any of these topics.
Submission
For abstract submission, please include the name(s) of the author(s), affiliations, and a contact e-mail address in the first lines of the body of the message. Texts should be written in English. Since the number of participants is limited to 40, registration will be restricted and based on the scientific quality of the submitted abstract. Authors are invited to present their work in discussion groups or poster sessions at the summer school.
All details can be viewed at the summer school website
Important dates
Deadline for the application is the 2nd of May, 2007!
Notification of acceptance May 21st, 2007
Summer school September 16th-21st,2007
Registration
The number of participants is limited to 40.
There will be no registration fee. Participants will have to pay for lodging and board. We are currently trying to get further funding for participants.
Invited speakers are:
Monica Baciu (LPNC, UPMF, Grenoble)
Grzegorz Dogil (Stuttgart university)
Hélène Loevenbruck (ICP/Gipsa-lab, CNRS, Grenoble)
Marc Sato (CRLMB, McGill university, Montréal)
Jean-Luc Schwartz (ICP/Gipsa-lab, CNRS, Grenoble)
Christophe Pallier (INSERM U562, Gif sur Yvette)
Georg Meyer (School of psychology, university of Liverpool)
Bernd Kröger (UK Aachen)
Organizers
Susanne Fuchs (ZAS, Berlin)
Hélène Loevenbruck (ICP, GIPSA-lab, Grenoble)
Daniel Pape (ZAS, Berlin)
Pascal Perrier (ICP, GIPSA-lab, Grenoble)

Studentships available for 2006/7 at the Department of Computer Science
The University of Sheffield - UK

One-Year MSc in HUMAN LANGUAGE TECHNOLOGY
The Sheffield MSc in Human Language Technology has been carefully tailored to meet the demand for graduates with the highly-specialised multi-disciplinary skills that are required in HLT, both as practitioners in the development of HLT applications and as researchers into the advanced capabilities required for next-generation HLT systems. The course provides a balanced programme of instruction across a range of relevant disciplines including speech technology, natural language processing and dialogue systems.
The programme is taught in a research-led environment. This means that you will study the most advanced theories and techniques in the field, and also have the opportunity to use state- of-the-art software tools. You will also have opportunities to engage in research-level activity through in-depth exploration of chosen topics and through your dissertation.
Graduates from this course are highly valued in industry, commerce and academia. The programme is also an excellent introduction to the substantial research opportunities for doctoral-level study in HLT.
A number of studentships are available, on a competitive basis, to suitably qualified applicants. These awards pay a stipend in addition to the course fees.
See further details of the course
Information on how to apply

top

BOOKS, DATABASES, SOFTWARES

Databases

HIWIRE database
We would like to draw your attention to the Interspeech 2007 special session "Novel techniques for the NATO non-native Air Traffic Control and HIWIRE cockpit databases"
http://www.interspeech2007.org/Technical/nato_atc.php
that we are co-organizing. For this special session we make available (free of charge) the cockpit database, along with training and testing HTK scripts. Our goal is to investigate feature extraction, acoustic modelling and adaptation algorithms for the problem of (hands-free) speech recognition in the cockpit. A description of the task, database and ordering information can be found at the website of the project We hope that you will be able to participate to this special session.
Alex Potamianos, TUC
Thibaut Ehrette, Thales Research
Dominique Fohr, LORIA
Petros Maragos, NTUA
Marco Matassoni, ITC-IRST
Jose Segura, UGR

- Language Resources Catalogue - Update
ELRA is happy to announce that 3 new Speech Related Resources are now available in its catalogue. Moreover, we are pleased to announce that years 2005 and 2006 from the Text Corpus of "Le Monde" (ELRA-W0015) are now available.
*ELRA-S0235 LC-STAR Hebrew (Israel) phonetic lexicon *The LC-STAR Hebrew (Israel) phonetic lexicon comprises 109,580 words, including a set of 62,431 common words, a set of 47,149 proper names (including person names, family names, cities, streets, companies and brand names) and a list of 8,677 special application words. The lexicon is provided in XML format and includes phonetic transcriptions in SAMPA. More information
*ELRA-S0236 LC-STAR English-Hebrew (Israel) Bilingual Aligned Phrasal lexicon *The LC-STAR English-Hebrew (Israel) Bilingual Aligned Phrasal lexicon comprises 10,520 phrases from the tourist domain. It is based on a list of short sentences obtained by translation from US-English 10,449 phrasal corpus. The lexicon is provided in XML format. More information
*ELRA-S0237 LC-STAR US English phonetic lexicon *The LC-STAR US English phonetic lexicon comprises 102,310 words, including a set of 51,119 common words, a set of 51,111 proper names (including person names, family names, cities, streets, companies and brand names) and a list of 6,807 special application words. The lexicon is provided in XML format and includes phonetic transcriptions in SAMPA. More information
*ELRA-W0015 Text corpus of "Le Monde" *Corpus from "Le Monde" newspaper. Years 1987 to 2002 are available in an ASCII text format. Years 2003 to 2006 are available in .XML format. Each month consists of some 10 MB of data (circa 120 MB per year). More information
For more information on the catalogue, please contact Valérie Mapelli
Our on-line catalogue has moved to the following address. Please update your bookmarks.

Books

Human Communication Disorders/ Speech therapy
This interesting series can be listed on Wiley website

Incurses em torno do ritmo da fala
Author: ( Plinio A. Barbosa
Publisher: Pontes Editores (city: Campinas)
Year: 2006 (released 11/24/2006)
(In Portuguese, abstract attached.) Website

Speech Quality of VoIP: Assessment and Prediction
Author: Alexander Raake
Publisher: John Wiley & Sons, UK-Chichester, September 2006
Website

Self-Organization in the Evolution of Speech, Studies in the Evolution of Language
Author: Pierre-Yves Oudeyer
Publisher:Oxford University Press
Website

Speech Recognition Over Digital Channels
Authors: Antonio M. Peinado and Jose C. Segura
Publisher: Wiley, July 2006
Website

Multilingual Speech Processing
Editors: Tanja Schultz and Katrin Kirchhoff ,
Elsevier Academic Press, April 2006
Website

Reconnaissance automatique de la parole: Du signal a l'interpretation
Authors: Jean-Paul Haton
Christophe Cerisara
Dominique Fohr
Yves Laprie
Kamel Smaili
392 Pages
Publisher: Dunod

top

JOB 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)

Language Technology and Cognitive Systems
International Research Training Group
Saarbruecken - Edinburgh

The International Research Training Group (IRTG) in Language Technology and Cognitive Systems announces the availability of up to 5 PhD positions beginning Oct. 1st, 2007. Established in 2001, the IRTG is a collaborative PhD programme between Saarland University and the University of Edinburgh - two leading institutions in the fields of computational linguistics, cognitive science and artificial intelligence. The participating institutes in Saarbruecken are Computational Linguistics, Computer Science, Phonetics, Psycholinguistics, and Speech and Signal Processing. In Edinburgh, the College is supported by the School of Informatics, the School of Philosophy, Psychology and Language Sciences, and the Human Communication Research Centre.
Each scholarship is funded for two years, normally extendable for a  third year. Doctoral degrees may be obtained in computational linguistics (also covers psycholinguistics), phonetics, or informatics,
from Saarland University. The official language of the programme is English. Distinguishing features of the cooperation include:
- Joint supervision of dissertations by lecturers from Saarbruecken and Edinburgh
- A six month research stay in Edinburgh
- An intensive research exchange programme between Saarbruecken and Edinburgh
The college focuses on the computational and cognitive foundations of human language processing, particularly emphasizing the following research areas:
- Cognitive and computational investigation of speech and language understanding
- Knowledge representation, inference, the lexicon, and ontologies
- Data-intensive and statistical language processing
- Dialogue, language generation and human language processing
For more information about the graduate college, including current staff and students, prospective applicants are encouraged to visit our website
The scholarship currently provides EURO 1468 per month. Additional compensation includes family allowance (where applicable), travel funding, support for carrying out experiments, and an additional monthly allowance for the duration of the stay in Edinburgh. Applicants should hold a strong university degree equivalent to the German Diplom or Magister (e.g. Master's level) in a relevant discipline, and should not be more than 28 years of age. Women and international students are particularly encouraged to apply.
Deadline for applications: 30 April, 2007
Applications should be sent electronically
To be fully considered, please include:
- a curriculum vitae indicating degrees obtained, disciplines covered (e.g. list of courses ), publications, and other relevant experience
- a sample of written work (e.g. research paper, or dissertation, preferably in English)
- copies of high school and university certificates, and transcripts
- two references (e-mailed to the speaker by the reference authors before the deadline)
- a statement of research interests, previous knowledge and activities in any of the relevant research areas.
In case an application can only be submitted by regular post, it should be sent to:
Prof. Dr. Matthew Crocker (Speaker)
Department of Computational Linguistics
Saarland University
P.O. Box 15 11 50
D-66041 Saarbruecken Germany
Tel. +49 (0)681 302-6560 - Fax +49 (0)681 302-6561
E-mail
Internet

Maitre de Conference en Reconnaissance et Comprehension de la Parole -Universite Rene Descartes Paris 5
Un poste de MCF (27 MCF 1616) en Reconnaissance et compréhension de la parole est à pourvoir à l'université René Descartes-Paris 5 (UFR de Mathématiques et Informatique) avec le profil suivant en recherche et en enseignement :
RECHERCHE
Le CRIP5 est un laboratoire d'informatique, avec des axes de recherche spécifiques et une production de niveau international. C’est aussi un laboratoire de recherche appliquée, résolument orienté vers les domaines qui font l’originalité de l’université Paris 5 (sciences de la vie et sciences humaines). L’équipe Dialogue et Indexation (Diadex) s’intéresse à tous les domaines de recherche de la reconnaissance et de la compréhension de la parole (modèles acoustiques et linguistiques, stratégies de décodage et optimisation, modèles de langage de genre, planification du dialogue, grammaires formelles pour le langage naturel, …). Le nouveau maître de conférences devra s’intégrer dans l’équipe Diadex et avoir une expérience solide dans un ou plusieurs des domaines précédemment cités. Il devra s’impliquer dans les différents projets de l’équipe et participer à l’encadrement des étudiants du master recherche.
ENSEIGNEMENT
Le nouveau Maître de Conférences prendra en main les enseignements de programmation et participera activement à leur organisation tout au long des trois années de la Licence Mathématiques Informatique et Applications. Il participera également au développement du parcours « Parole et Communication Homme-Machine » du Master Recherche MISV, spécialité Informatique
Contact : Prof. Marie-Jose Caraty
Website

PHD RESEARCH FELLOWSHIPS RESEARCH INTERNSHIPS at THE UNIVERSITY of TRENTO ITALY

The Adaptive Multimodal Information and Interfaces Research Lab at University of Trento (Italy) has several openings.
Topic areas: Automatic Speech Recognition, Spoken Language Understanding, Dialogue Modeling, Machine Learning
The Research Lab: The Adaptive Multimodal Information and Interface (AMI2) research lab pursues excellence research in next-generation interfaces for human-machine and human-human communication. The Ph.D. research fellowships are funded by the prestigious Marie Curie Excellence grant awarded by the European Commission for cutting edge and interdisciplinary research. There have been many advances in the field of speech recognition and text-to-speech synthesis in the last decade. However, human-machine conversational system are still severely limited in the ability to understand and interact. The general research challenge of the project is to investigate innovative multimodal interfaces that will be adaptive, affective, usable and knowledgeable.
The Adaptive Multimodal Information and Interface (AMI2) research lab pursues excellence research in next-generation interfaces for human-machine and human-human communication. The AMI2 lab has a state-of-the-art technology infrastructure and collaborations with premiere international research centers and industry research labs.
The Ph.D. Researchers: The Ph.D. researchers should have a background (Laurea, MA or equivalent) in speech/text processing, machine learning, computational linguistics, computer science or cognitive science, and strong academic records. The Ph.D. positions are funded for 3 years. The Ph.D. students will be able to participate in the international Ph.D. program of the ICT Department . The applicants should be fluent in English. The Italian language competence is optional and applicants are encouraged to acquire this skill during their Ph.D. studies. The applicants should have good programming skills in conventional programming languages (C++/Java//Perl/Python) and experience in rule-based, logic or functional programming.
The University of Trento is an equal opportunity employer. Interested applicants should submit their CV along with their statement of research interest and reference letters before March 31, 2007.
Applications for research internships are possible at any time. It is recommended to mailto: contact us before submitting a formal application:
Prof. Ing. Giuseppe Riccardi http://www.dit.unitn.it/~riccardi
Dr. Sebastian Varges http://www.dit.unitn.it/~varges

Two positions at the University of Wolverhampton

Professor or Reader in Computational Linguistics/Natural Language Processing
Reference number: B4553
Application deadline: 9 March 2007
We are seeking a Professor or Reader (depending on track record and qualifications) in Computational Linguistics/Natural Language Processing with an outstanding research record as evidenced by publications in leading journals and successful grant applications. Commercial experience or links with industry would be an advantage as would teaching experience and experience in mentoring less experienced researchers. Duties include collaborative research and producing first class research/publications, grant applications, PhD student/researcher supervision and teaching. This is a permanent position.
Applications should include a completed application form, CV, and covering letter explaining your reasons for applying for the position, including the level of appointment, and giving details of your research interests/experience and background. You should also give the names of three referees who can comment on your suitability for appointment at the appropriate level. The interviews are scheduled to take place in April, with the starting date as soon as possible thereafter.
Senior Research Fellow in Computational Linguistics/Natural Language Processing
Applications closing date: 9 March 2007
Reference: A4686
We are seeking a Senior Researcher in Computational Linguistics/Natural Language Processing with an excellent research record as evidenced by publications in reputable journals. Project experience or links with industry would be an advantage as would teaching or PhD supervision experience. Duties include collaborative research and producing first class research/publications, involvement in grant applications, and some teaching and student supervision. This appointment is for two years in the first instance. Applications should include a completed application form, CV, and covering letter in which you explain why you have applied for the position and give details of your research interests/experience and background. The covering letter should also present a research plan (about 1,000 words) which you would like to pursue. The interviews are scheduled to take place in April with the starting date as soon as possible thereafter.
For informal enquires, please contact Prof. Ruslan Mitkov. Applications to be sent to Personnel. If emailed, please cc to R.Mitkov@wlv.ac.uk. For an application form, contact:
Personnel Services Department,
University of Wolverhampton,
Molineux Street,
Wolverhampton WV1 1SB
Phone +44 (0) 1902 321049 (ansaphone), and quoting the reference number.
For hearing impaired candidates our Minicom number is +44 (0) 1902 321249.
Email address
The application form can also be downloaded
* In line with the University pay modernisation programme, salaries may be subject to review.
Established by Prof. Mitkov in 1997, the Research Group in Computational Linguistics is a highly successful one, delivering cutting-edge research in a number of NLP areas such as anaphora resolution, automatic summarisation, question answering, lexical knowledge acquisition, text categorisation, named entity recognition, information extraction, corpus construction and annotation, automatic terminology processing, multilingual processing, and multiple-choice question generation. To a large extent, this research has been undertaken in projects funded by major UK funding bodies and commercial partners.

Opening on Speech recognition at Telefonica, Barcelona (Spain)

The speech Technology Group at Telefonica Investigacion y Desarrollo (TID) is looking for a highly qualified candidate for an engineering position on speech recognition and related technologies.
The selected person will become a part of a multidisciplinary team of young highly motivated people in an objective driven, friendly atmosphere located in a central area of Barcelona (Spain).
Minimum requirements ar:
Degree in Computer Science /Electrical Engineering/Computational Linguistics or similar with 2+ years of experience (Ph.D. preferred) on speech technology.
Good knowledge of speech recognition and speech synthesis.
Proven programming expertise in C++ and Java
Good level of English (required) and some knowledge of Spanish (preferred)
High motivation and teamwork spirit
Salary depending on the experience and value of the applicant
Starting date as soon as possible
The speech technology group is a well established group within TID with more than 15 years of experience in research and development of technology for internal use of Telefonica group as well as outside organizations. It is also a very active partner in many National and European projects. TID is the research and development company inside the Telefonica group, currently one of the biggest Telecom companies. It is the biggest private research center in Spain in number of employees and available resources.
Please send your resume and contact information to
Sonia Tejero
Tlf: +34 93 365 3024

Sound to Sense: 18 Fellowships in speech research

Sound to Sense (S2S) is a Marie Curie Research Training Network involving collaborative speech research amongst 13 universities in 10 countries. 18 Training Fellowships are available, of which 12 are predoctoral and 6 postdoctoral (or equivalent experience). Most but not all are planned to start in September or October 2007.
A research training network’s primary aim is to support and train young researchers in professional and inter-disciplinary scientific skills that will equip them for careers in research. S2S’s scientific focus is on cross-disciplinary methods for modelling speech recognition by humans and machines. Distinctive aspects of our approach include emphasis on richly-informed phonetic models that emphasize communicative function of utterances, multilingual databases, multiple time domain analyses, hybrid episodic-abstract computational models, and applications and testing in adverse listening conditions and foreign language learning.
Eleven projects are planned. Each can be flexibly tailored to match the Fellows’ backgrounds, research interests, and professional development needs, and will fall into one of four broad themes.
1: Multilinguistic and comparative research on Fine Phonetic Detail (4 projects)
2: Imperfect knowledge/imperfect signal (2 projects)
3: Beyond short units of speech (2 projects)
4: Exemplars and abstraction (3 projects)
The institutions and senior scientists involved with S2S are as follows:
* University of Cambridge, UK (S. Hawkins (Coordinator), M. Ford, M. Miozzo, D. Norris. B. Post)
* Katholieke Universiteit, Leuven, Belgium (D. Van Compernolle, H. Van Hamme, K. Demuynck)
* Charles University, Prague, Czech Republic (Z. Palková, T. Dub?da, J. Volín)
* University of Provence, Aix-en-Provence, France (N. Nguyen, M. d’Imperio, C. Meunier)
* University Federico II, Naples, Italy (F. Cutugno, A. Corazza)
* Radboud University, Nijmegen, The Netherlands (L. ten Bosch, H. Baayen, M. Ernestus, C. Gussenhoven, H. Strik)
* Norwegian University of Science and Technology (NTNU), Trondheim, Norway (W. van Dommelen, M. Johnsen, J. Koreman, T. Svendsen)
* Technical University of Cluj-Napoca, Romania (M. Giurgiu)
* University of the Basque Country, Vitoria, Spain (M-L. Garcia Lecumberri, J. Cenoz)
* University of Geneva, Switzerland (U. Frauenfelder)
* University of Bristol, UK (S. Mattys, J. Bowers)
* University of Sheffield, UK (M. Cooke, J. Barker, G. Brown, S. Howard, R. Moore, B. Wells)
* University of York, UK. (R. Ogden, G. Gaskell, J. Local)
Successful applicants will normally have a degree in psychology, computer science, engineering, linguistics, phonetics, or related disciplines, and want to acquire expertise in one or more of the others.
Positions are open until filled, although applications before 1 May 2007 are recommended for starting in October 2007.
Further details are available from the web about: + the research network and how to apply: http://www.ling.cam.ac.uk/s2s/s2sJobAd.pdf (92 kB) + the research projects: http://www.ling.cam.ac.uk/s2s/s2sProjects.pdf (328 kB).
Post-doc in Japan - Machine learning, kernel machines, computational statistics, Bayesian statistics, or multimodal processing

DeadLine: 01/05/2007
Website
The Institute of Statistical Mathematics (ISM)
Research Organization of Information and Systems (ROIS)
Postdoctoral position: Applicants are invited to apply to the Transdisciplinary Research Integration Center, ISM/ROIS. ISM is one of member institutes of ROIS along with the National Institute of Informatics, the National Institute of Genetics and the National Institute of Polar Research. The ISM mission includes promoting statistical science and developing an innovative methodology for approaching complex problems related to life science, earth science, environmental science and human sciences from the view point of information and systems (http://www.ism.ac.jp/index_e.html). The position will start as soon as possible after 1 April 2007. The postdoctoral researcher will work on the following project “Discovery of Invariants in Multimodal Data” (http://www.ism.ac.jp/~tmatsui/kinou2_p4/index-en.html). The initial contract is one-year long but could be extended up to three years.
Field of work: Machine learning, kernel machines, computational statistics, Bayesian statistics, or multimodal processing
Project description: Multimodal data available to us through the Internet and other electronic media are explosively increasing both in number and in variety. To handle such massive data for various purposes, new technologies need to be developed. With this in mind, we have started investigating a new methodology that allows us to discover from multimodal data the information relevant to the purpose at hand (which is referred to as “invariants”). To achieve this goal, we will study several qualitatively different problems from different research areas, in which multimodal data play a central role (e.g., visual/audio/text processing, cognitive science, auditory perception and robotics). The problems are to be tackled with some of the recently developed inductive learning machines including automatic model selection mechanism (e.g., Penalized Logistic Regression Machines and Support Vector Machines). The results will be analyzed in order to establish a new methodology for discovery of invarian! ts, which will be applicable to problems across different areas of study.
Job description: The successful candidate will support and coordinate our efforts in the area of investigation of methods for discovery of invariants with multimodal data.
Requirements: Applicants should have a PhD and some knowledge of machine learning and statistics. Applicants must be able to program (C/C++ and Matlab knowledge is an advantage but not requirement) and must also have experience with statistical data analysis.
Payment: The salary will be in the range of 4,500,000 yen 6,000,000 yen (before tax and insurance).
Application: Applicants should send their CV, including a list of publications and the names of two potential referees.
Contact: Prof. Tomoko Matsui
Tel: +1 604 822 9662 (until 9 March 2007 (Vancouver, Canada)
+81 3 5421 8769 (from 10 March 2007 (Japan)

Professorship at the University Paris Sud, Orsay, France

A professorship is available in language processing (speech and text processing, automatic translation, statistical learning). The new professor will reinforce teaching and carry out research in the general area of computational linguistics. The candidate will also be expected to participate in teaching core subjects in computer science. The research will be carried out at LIMSI-CNRS.
Announcement:
http://www.limsi.fr/Formation/prop_postes/ProfTAL.html
Contacts:
Philippe Tarroux +33.1.69.85.80.86
Jean-Luc Gauvain +33.1.69.85.80.63

Postdoctoral researcher at IRISA Rennes France

We are hiring a postdoctoral researcher on the statistical modeling of multichannel audio, applied to speaker segmentation and separation (full subject below). The successful candidate will work under the supervision of Drs. Guillaume Gravier and Emmanuel Vincent, in the METISS group at IRISA, which possesses a newly-equipped room dedicated to the exploration of future meeting environments.
Prospective candidates should have a background in multichannel signal processing or in speech processing and hold a PhD for less than one year or being about to obtain one. Informal enquiries may be made to Emmanuel Vincent or Guillaume Gravier . This appointment is for 2 years, starting summer or fall 2007. Salary will be at 28000 euros per annum. Applications must be submitted online before march 31st at http://www.inria.fr/travailler/opportunites/postdoc/postdoc.en.html
Description of the work
Joint statistical modeling of spectral, temporal and spatial audio features, applied to speaker segmentation and separation Most audio signals represent complex sound scenes consisting of several overlapping sources (speakers, natural sounds, musical instruments). These sources are usually located at different spatial positions and exhibit different spectro-temporal characteristics. The processing of such documents involves several challenging tasks, such as the separation, the segmentation and more generally the description of each source. Existing description algorithms are mostly designed for one-microphone recordings and rely on statistical modeling of spectral features. Yet, in many application environments, multiple microphones are available thus providing valuable spatial information. Beamforming algorithms are then typically employed to determine at each instant the number of sources and their locations based on spatial features. These algorithms can improve the detection of overlapping sources. However their robustness decreases for small microphone arrays or with moving sources. The goal of this project is to define a unified statistical modeling framework for the joint exploitation of spectral, temporal and spatial information in multichannel audio signals. Dynamic state-based models offer a promising approach for the description of some extracted spectral and spatial features as a function of some hidden states associated with different sources and positions. A first stage of the project could consist of extending the state-of-the-art one-microphone segmentation model developed in our lab (based on GMMs) by incorporating spatial features obtained from classical source localization and separation techniques (e.g. ICA, DUET, beamforming). The proposed framework will be primarily applied to speaker segmentation and separation, which is the task of finding out the structure of a speech recording according to the question "who spoke when and where" and to extract the signal of each speaker. The results will be evaluated on meeting data recorded by small microphone arrays. Data from the NIST meeting evaluation will be used along with data recorded at our lab in a room dedicated to the exploration of future meeting environments.
Web

Postdoctoral position- INRIA-LORIA Nancy France

Objective
Despite recent progresses achieved in speech synthesis, it is still very difficult to modify the characteristics linked to the speaker since signals are synthesized by concatenating sounds uttered by a given speaker. It is thus almost impossible to modify acoustic cues of sounds as well as characteristics linked to the speaker.
The objective of the postdoc is to elaborate copy synthesis algorithms that enable a speech signal to be reproduced as faithfully as possible while offering the possibility of modifying acoustic cues. For this reason this postdoctoral work will rely on the utilization of a formant synthesizer derived from that proposed by Klatt[1]. Synthesis thus rests on the filtering by a system of resonators (representing formants) of a sound source, periodic for the voiced sounds as vowels, aperiodic (a noise) for unvoiced sounds as fricatives phonemes.
Work
The work will consist of adapting the synthesizer so that it does lend itself to copy synthesis as well as possible and to develop algorithms to optimizing source and formant parameters.
In order to copy speech sufficiently finely it is necessary to adjust formant and source parameters precisely. The LF source model proposed by Fant and Liljencrants [2] is sufficiently versatile to approximate a natural speech source. The optimization of the four parameters was the subject of a number of works in the case where the vocal tract filter and source are estimated jointly [3,4] or when the source signal is known [5]. The specificity of copy synthesis is that the filter of the vocal tract is only roughly approximated by formants hypothesised and that the ratio of noise in the source has also to be adjusted for each of the formants.
Resonators of a formant synthesizer can be organized in cascade or in parallel. Only the second solution is usable in the case of copy synthesis because it enables formants to be adjusted independently [6]. The frequency, amplitude and bandwidth of each formant have to be specified. One important advantage of the parallel architecture is that it is possible to adjust only amplitude by setting the bandwidth to a default value once the formant frequency is known. The second aspect of the work will be on the elaboration of an algorithm to adjust amplitudes and frequencies. The adjustment of amplitudes must be synchronized on source periods in order to capture fast variations of amplitude, and that of formant frequencies will rest upon the automatic formant tracking previously developed [7]. Improvements will be about the choice of the formant number so as to increase the closeness of the speech copied with respect to the original signal.
The two aspects have been presented independently to simplify the presentation of the work. To a certain extent only they also can be addressed independently. However, it is clear that the improvement of the synthesis quality will be all the better since interactions between these two aspects will have been considered together.
The Parole team mainly works on automatic speech recognition and speech analysis. In the domain of analysis a number of algorithms have been developed (F0 detection, formant tracking, pitch marking, copy synthesis...) and are available in WinSnoori software which already contains a series of tools for copy synthesis and which is developed by the team for several years.
Skill and profile
A good knowledge in speech analysis or in signal processing is required.
References
Copy synthesis tools of WinSnoori are presented here.
[1] D.H. Klatt, “Software for a cascade/parallel formant synthesizer”, J. Acoust. Soc. Amer., 67(3), p. 971-995, March 1980.
[2] G. Fant and J. Liljencrants, “A four parameter model of glottal flow”, STL, QPSR, 4, p. 1-13, 1985
[3] M. Frölich , D. Michaelis and H.W. Strube, “SIM-simultaneous inverse filtering and matching of a glottal flow model for acoustic speech signals”, J. Acoust. Soc. Amer., 115(1), p.337-351, 2003.
[4] D. Vincent, O. Rosec and T. Chonavel, “Estimation of LF glottal source parameters based on an ARX model”, Proc. of Interspeech, p. 333-336, Lisboa, Sep. 2005.
[5] J. Pérez and A. Bonafonte, “Automatic Voice-Source Parametrization of Natural Speech”, Proc. of Interspeech, Lisboa, Sep. 2005.
[6] W. J. Holmes, “Copy synthesis of female speech using the JSRU parallel formant synthesiser”, Proceedings of European Conference on Speech Technology, p. 513-516, Paris, France, Sep., 1989
[7] Y. Laprie, “A concurrent curve strategy for formant tracking”, Proc. of ICSLP, Jegu, Korea, Oct. 2004
Contact
Interested candidates are invited to contact Yves Laprie
Important information
This position is advertised in the framework of the national INRIA campaign for recruiting post-docs. It is a one year position, renewable, beginning fall 2007. The salary is 2,320€ gross per month.
Selection of candidates will be a two step process. A first selection for a candidate will be carried out internally by the PAROLE group. The selected candidate application will then be further processed for approval and funding by an INRIA committee.
Doctoral thesis less than one year old (May 2006) or being defended before end of 2007. If defence has not taken place yet, candidates must specify the tentative date and jury for the defence.
Useful link
Presentation of INRIA postdoctoral positions To apply(be patient, loading this link takes times…)

Research scientist- Speech Technology- Princeton, NJ, USA

Company Profile: Headquartered in Princeton, NJ, ETS (Educational Testing Service)is the world's premier educational measurement institution and a leader in educational research. As an innovator in developing achievement and occupational tests for clients in business, education, and government, we are determined to advance educational excellence for the communities we serve.
Job Description: ETS Research & Development has a Research Scientist opening in the Automated Scoring and Natural Language Processing Group. This group conducts research focusing on the development of new capabilities in automated scoring and NLP-based analysis and evaluation systems, which are used to improve assessments, learning tools and test development practices for diverse groups of users that include K-12 students, college students, English Language Learners and lifelong learners. The Research Scientist position involves applying scientific, technical and software engineering skills to designing and conducting research studies and developing capabilities in support of educational products and services. The job is a full-time job.
Required qualifications
· A Ph.D. in Natural Language Processing, Computational Linguistics, Computer Science, or Electrical Engineering with a focus on speech technology, particularly speech recognition. Knowledge of linguistics is a plus.
· Evidence of at least three years of independent substantive research experience and/or experience in developing and deploying speech technology capabilities, preferably in educational environments.
· Demonstrable contributions to new and/or modified theories of speech processing and their implementation in automated systems.
· Practical expertise with speech recognition systems and fluency in at least one major programming language (e.g., Java, Perl, C/C++, Python).
· Three years of independent substantive research experience and/or experience in developing and deploying speech technology capabilities, preferably in educational environments.
How to apply
Please send copy of your resume, along with cover letter stating salary requirements and job #2965, to e-mail
ETS offers competitive salaries, outstanding benefits, a stimulating work environment, and attractive growth potential. ETS is an Equal Opportunity, Affirmative Action Employer.
Web site

Research Associate in Spoken Language Processing
Department of Computer Science, University of Sheffield, UK

Ref No: PR2783
Closing Date: 3rd April, 2007
Summary
Following the award of an EU-funded Framework 6 project called ACORNS, we are looking for an innovative researcher to join a well-established and internationally recognised, interdisciplinary research group. Sheffield is responsible for research in the area of memory organisation and access, and the successful applicant is expected to take a lead role in the investigation of computational models for working memory, attention mechanisms, episodic vs. semantic memory and sensorimotor integration. Ideal candidates, in addition to having excellent teamwork and communication skills, should have high-calibre technical skills in one or more of the following: spoken language processing, speech technology, artificial intelligence or machine learning. This post is available immediately until 30 November 2009.>br> Full Post Details
Application Pack
On-Line Application Form website and enter vacancy number 3006866.

Software Engineer Position at Be Vocal, Mountain View, CA,USA

We are currently looking for a Software Engineer with previous exposure to Speech, to work in our Speech and Natural Language Technology group. This group’s mission is to be the center of excellence for speech and natural language technologies within BeVocal. Responsibilities include assisting in the development of internal tools and processes for building Natural Language based speech applications as well as on ongoing infrastructure/product improvements. The successful candidate must be able to take direction from senior members of the team and will also be given the opportunity to make original contributions to new and existing technologies during the application development process. As such, you must be highly motivated and have the ability to work well independently in addition to working as a team.
Responsibilities
* Develop and maintain speech recognition/NLP tools and supporting infrastructure
* Develop and enhance component speech grammars
* Work on innovative solutions to improve overall Speech/NL performance across BeVocal’s deployments.
Requirements
* BS in Computer Science, Electrical Engineering or Linguistics, an MS is a preferred.
* 2-5 years of software development experience in Perl, Java, C/C++. A willingness and ability to pick up additional software languages as needed is essential.
* Exposure or experience with speech recognition/pattern recognition either from an academic environment or directly related work experience.
* Experience working as part of a world-class speech and language group is highly desirable.
* Experience building natural language applications is preferred.
* Experience building LVCSR speech recognition systems is a plus.
For immediate consideration, please send your resume by email and include "Software Engineer, Speech" in the subject line of your email. Principals only please (no 3rd parties or agencies). Contact for details
BeVocal's policy is to comply with all applicable laws and to provide equal employment opportunity for all applicants and employees without regard to non-job-related factors such as race, color, religion, sex, national origin, ancestry, age, disability, veteran status, marital status or sexual orientation. This policy applies to all areas of employment, including recruitment, hiring, training, promotion, compensation, benefits, transfer, and social and recreational programs.

Postdoctoral Fellow -- Speech Synthesis- Alfred I. Dupont Hospital for Children, Wilmington, DE

The Alfred I. duPont Hospital for Children in Wilmington, DE has an immediate opening for a Postdoctoral Fellow in Speech Synthesis in the Speech Research Laboratory, within the Department of Biomedical Research. The ideal candidate will have a Ph.D. in Computer Science, Linguistics, Psychology, or a related field, demonstrated experience in data-based speech synthesis techniques, and an interest in modeling prosody, particularly intonation, in speech synthesis systems. The primary responsibilities for this position include: Developing a model for intonation that can be trained on and capture the important talker-specific features of an individual's speech while also representing phonologically motivated f0 characteristics; implementing the intonation model for the ModelTalker TTS system; and assisting in the creation of unit concatenation voices for the ModelTalker TTS system. A Ph.D. in Linguistics, Computer Science, Psychology, or closely related field with demonstrated knowledge of and experience in concatenative speech synthesis techniques, speech analysis techniques, and acoustic phonetics is required. Computer programming experience with C or C++, knowledge of additional languages is a plus. Experience with Unix/Linux and Windows operating systems is essential.
This is a two-year grant-funded position. For more information, email Dr Timothy Bunnel or call at (302) 651-6835. Applicants may also post their resume on-line at www.nemours.org or send resume with salary requirements to Dr. Timothy Bunnell, Department of Biomedical Research, Alfred I. duPont Hospital for Children, P.O. Box 269, Wilmington, DE 19899.

Position at Saybot in China

Job title: Speech Scientist
Location: China (Beijing or Shanghai)
Saybot develops software technology and curricula for learning spoken english. Since 2005, we have been building software which features state-of-the-art speech technologies and innovative interactive lessons to help users practice speaking English. We are currently looking for talented speech scientists to help strengthen our R&D team and to develop our next-generation products. Successful candidates would have proven excellence and good work ethics in academic or industry context and demonstrated creativity in building speech systems with revolutionary designs.
* MS/PhD degree in speech technology (or related).
* Expertise in at least one of the following areas and basic knowledge of the others:
o acoustic model training,
o speaker adaptation,
o natural language understanding,
o prosody analysis,
o embedded recognizers.
* Excellent programming skills in both object-oriented languages (C++, C# or Java) and scripting (Perl or Python).
* Good knowledge and experience in at least one commonly used recognizer (HTK, Sphinx, Nuance...).
* Excellent communication skills in written and oral English.
* Experience in machine translation is a plus.
* Experience in VoIP integration is a plus.
* Experience in language teaching is a plus.

Contact: Sylvain Chevalier

2 Positions in Research and Development in "Audio description and indexing" at IRCAM-Paris

PRESENTATION OF THE SAMPLE ORCHESTRATOR PROJECT:
The goal of the Sample Orchestrator project is to develop and test new applications for managing and manipulating sound samples based on audio content. On the one hand the commercial availability of large databases of sound samples available on various supports (CD, DVD, online), are currently limited in their applications (synthesizers by sampling). On the other hand, recent scientific and technological development in audio indexing and database management allow the development of new musical functions: database management based on audio content, audio processing driven by audio content, development of orchestration tools.
TASKS:
Two positions are available from April 15th 2007 within the "Equipe Analyse/Synthese" of Ircam for (each) a 12 months total duration (possibility of extending the contracts). The main tasks to be done for the research and development positions are:
- Research and development of new audio features and algorithms for the description of instrumental, percussive and FX sounds.
- Research and development of new audio features and algorithms for the morphological description of sounds
- Research and development of new audio features and algorithms for sounds containing "loops"
- Research and development of algorithms for automatic audio indexing
- Research and development of algorithms for fast search by similarity in large databases
- Participation in the definition of the specification
- Participation in user evaluation and feedback
- Integration into the final application
RESEARCH POSITION:
REQUIRED EXPERIENCE AND COMPETENCE:
- High skills in Audio indexing and signal processing
- High skills in Matlab programming
- High productivity, methodical work, excellent programming style.
- Good knowledge of UNIX, Mac and Windows environments
SALARY:
According to background and experience.
DEVELOPMENT POSITION:
REQUIRED EXPERIENCE AND COMPETENCE:
- Skills in Audio indexing and signal processing
- High skills in C/C++ programming
- High productivity, methodical work, excellent programming style.
- Good knowledge of UNIX, Mac and Windows environments
SALARY:
According to background and experience.
EEC WORKING PAPERS:
In order to start immediately, the candidate should preferably have EEC citizenship or already own valid EEC working papers.
AVAILIBILITY:
The positions are available in the "Analysis/Synthesis" team in the R&D department from April 15th 2007 for (each) a duration of 12 months (possibility of extending the contracts).
TO APPLY:
Please send your resume with qualifications and informations addressing the above issues, preferably by email Xavier Rodet, Analyse/Synthese team manager).
or by fax at: (33 1) 44 78 15 40, care of Xavier.Rodet
or by surface mail to: Xavier Rodet, IRCAM, 1 Place Stravinsky, 75004 Paris.
IRCAM:
IRCAM is a leading non-profit organization dedicated to musical production, R&D and education in acoustics and music, located in the center of Paris (France), next to the Pompidou Center. It hosts composers, researchers and students from many countries cooperating in contemporary music production, scientific and applied research. The main topics addressed in its R&D department are acoustics, psychoacoustics, audio synthesis and processing, computer aided composition, user interfaces, real time systems. Detailed activities of IRCAM and its groups are presented on our WWW server.

RESEARCH AND DEVELOPMENT POSITION IN "AUDIO CONTENT ACCESS" at IRCAM (Paris)

PRESENTATION OF THE MUSICDISCOVER PROJECT :
The goal of the MusicDiscover project is to give access to the contents of musical audios recordings (as it is the case, for example, for texts), i.e. to a structured description, as complete as possible, of the recordings: melody, genre/style, rate/rhythm, instrumentation, musical structure, harmony, etc. The principal objective is thus to develop and evaluate means directed towards the contents, which include techniques and tools for analysis, indexing, representation and search for information. These means will make it possible to build and use such a structured description. This project of the ACI "Masses of Data" is carried out in collaboration between Ircam (Paris), Get-Telecom (Paris) and the LIRIS (Lyon) since October 2004. The principal lines of research are :
- Rhythmic analysis and detection of ruptures
- Recognition of musical instruments and indexing
- Source Separation
- Structured Description
- Research of music by similarity
- Recognition of musical titles
- Classification of musical titles in genre and emotion.
The available position relates to the construction and the use of the Structured Description in collaboration with the other lines of research.
DEVELOPMENTS TASKS:
A position is available from December 1st 2006 within the "Equipe Analyse/Synthese" of Ircam for a 9 months total duration. The contents of work are as follows:
- Participation in the design of a Structured Description
- Software development for construction and use of Structured Descriptions
- Participation in the definition and development of the graphic interface
- Participation in the evaluations
REQUIRED EXPERIENCE AND COMPETENCE:
- Experience of research in Audio Indexing and signal processing
- Experience in Flash, C and C++ and Matlab programming.
- High productivity, methodical work, excellent programming style.
- Good knowledge of UNIX and Windows environments.
AVAILABILITY :
- The position is available in the "Analysis/Synthesis" team in the R&D department from November 1st 2006 for a duration of 9 months.
EEC WORKING PAPERS :
- In order to start immediately, the candidate should preferably have EEC citizenship or already own valid EEC working papers.
SALARY:
- According to background and experience.
TO APPLY:
- Please send your resume with qualifications and informations adressing the above issues, preferably by email to Xavier Rodet, Analyse/Synthese team manager.
or by fax at: (33 1) 44 78 15 40, care of Xavier.Rodet
or by surface mail to: Xavier Rodet, IRCAM, 1 Place Stravinsky, 75004 Paris.
Introducing IRCAM
IRCAM is a leading non-profit organization dedicated to musical production, R&D and education in acoustics and music, located in the center of Paris (France), next to the Pompidou Center. It hosts composers, researchers and students from many countries cooperating in contemporary music production, scientific and applied research. The main topics addressed in its R&D departement are acoustics, psychoacoustics, audio synthesis and processing, computer aided composition, user interfaces, real time systems. Detailed activities of IRCAM and its groups are presented on our WWW server

top

JOURNALS

CfP: IEEE SIGNAL PROCESSING MAGAZINE- Special Issue on Spoken Language Technology

The evolution of speech and language technologies over the past decade has spawned an exciting new research area known as Spoken Language Tech nology (SLT). Technological advances in SLT promise to provide ubiquit ous and personalized access to information, communication, and entertai nment services. For example, advances in natural language understanding and large vocabulary continuous speech recognition have resulted in a new generation of automated contact center services that offer callers the flexibility to speak their request naturally using their own words as opposed to the words dictated to them by the machine. Advances in ma chine translation technology have resulted in speech-to-speech translat ion products that offer multi-party multi-lingual communication. Advanc es in information search and data mining are providing the means to ext ract intelligence information from large corpora of speech data (e.g., TV programs, call center data) to help improve business operation and s earch for information rapidly without having to listen to conversations .
This special issue on Spoken Language Technology is motivated by the fi rst SLT workshop, Aruba, December 2006, jointly sponsored by IEEE and A CL (www.slt2006.org). The goal is to solicit tutorial articles with com prehensive surveys of important theories, algorithms, tools, and applic ations of SLT on existing and new commercial, academic and government a pplications. Prospective authors should submit a white paper summarizin g the motivation, the significance of the topic, brief history, and an outline of the content. Authors with accepted proposals will be invited to write a full manuscript.
Scope of topics:
Publications in the following areas are strongly encouraged
Spoken language understanding
Dialog management
Spoken language generation
Spoken document retrieval
Information extraction from speech
Question answering from speech
Spoken document summarization
Machine translation of spoken language
Speech data mining and search
Voice-based human computer interfaces
Spoken dialog systems, applications and standards
Multimodal processing, systems and standards
Machine learning for spoken language processing
Speech and language processing in the world wide web
Submission Procedure:
Prospective authors should submit their white papers to the web submiss ion system at http://www.ee.columbia.edu/spm according to the following timetable. The white papers should be three pages maximum
Important dates
White paper due: June 1, 2007
Invitation notification: July 1, 2007
Manuscript due: October 1, 2007
Acceptance Notification: December 1, 2007
Final Manuscript due: January 15, 2008
Publication date: May, 2008
Guest Editors:
Mazin Gilbert
AT&T Labs - Research
180 Park Avenue
Florham Park, NJ, 07932
Kevin Knight
University of Southern California
4676 Admiralty Way
Marina del Rey, CA 90292
Steve Young
Cambridge University
Trumpington Street
Cambridge, CB2 1PZ

Call for Papers- Special Issue of the IEEE Transactions on Audio, Speech and Language Processing on New Approaches to Statistical Speech and Text Processing

Dramatic advances in automatic speech recognition (ASR) technology in recent years has enabled serious growth in spoken language processing research, both for human-computer interaction and spoken document processing. The challenges of working with spoken language, including ASR errors and disfluencies, were major factors in the adoption of statistical techniques in the language processing community. Statistical methods now dominate many areas of text processing as well, enabled by growing collections of linguistic data resources and developments in machine learning. While transfer of methods from spoken- to written- language processing continues, advances in written-language processing also now have a significant impact on spoken-language processing. This issue seeks to highlight the cross-fertilization in speech and text processing by publishing novel statistical modeling and learning methods that span a variety of language processing applications.
We invite papers describing new approaches to statistical language processing of both spoken and written language. Submissions must not have been previously published, with the exception that substantial extensions of conference papers will be considered. Of particular interest are methods that transfer recent developments from text processing to speech processing and vice versa, but new methods in one domain are also welcome. Papers describing new strategies for integrating acoustic and linguistic cues in spoken language processing are also encouraged.
Topics of interest include:
- Unsupervised and semi-supervised learning
- Discriminative learning
- Transfer or adaptation to new domains
- Active learning
- Reinforcement learning
- Memory-based learning and neighborhood methods
- Novel statistical models
- Statistical methods for feature selection or transformation
Specific applications of interest include information extraction, question answering, text segmentation and classification, summarization, translation, language generation and spoken language dialogs. Papers that address component problems of these larger applications are also encouraged, including parsing, discourse analysis, and talker interaction analysis. The issue aims to cover a variety of applications as well as different statistical methods.
Submission procedure:
Prospective authors should prepare manuscripts according to the Information for Authors as published in any recent issue of the Transactions. Note that all rules will apply with regard to submission lengths, mandatory overlength page charges, and color charges. Manuscripts should be submitted electronically through the online IEEE manuscript submission system. When selecting a manuscript type, authors must click on "Special Issue of TASLP on New Approaches to Statistical Speech and Text Processing". Authors should follow the instructions for the IEEE Transactions Audio, Speech and Language Processing and indicate in the Comments to the Editor-in-Chief that the manuscript is submitted for publication in the Special Issue on New Approaches to Statistical Speech and Text Processing. We require a completed copyright form to be signed and faxed to +1-732-562-8905 at the time of submission. Please indicate the manuscript number on the top of the page.
Schedule:
Submission deadline: 15 June 2007
Notification of final acceptance: 15 December 2008
Final manuscript due: 1 February 2008
Publication date: May 2008
Guest Editors:
Dr. Bill Byrne Cambridge University, UK
Dr. Mark Johnson Brown University, USA
Dr. Lillian Lee Cornell University, USA
Dr. Steve Renals University of Edinburgh, UK

Call for papers for a special issue of Speech Communication on Iberian Languages

Iberian languages (henceforth IL) are amongst the most widely spoken languages in the world. Nowadays, 628 million people on virtually all continents have Spanish, Portuguese, Catalan, Basque, Galician, etc. as their official language. Consequently, important speech research centers and companies, both public and private, are focusing their interest on those languages. This effort has resulted in novel and generic approaches applicable to any language, as well as in the optimization of existing techniques or systems. It is worth highlighting that the community working on speech science and technology in IL speaking countries has already reached world-class level in many areas and has continuously increased in size in the last 15 years.
Speech technology proposed in the context of a non-Iberian language (e.g., English) may not be directly applicable to IL. All linguistic and paralinguistic dimensions, from phonetics to pragmatics, are amongst the features that certainly distinguish IL from others considered in speech science and technology research. As a result, original work and optimization of existing techniques and systems may be necessary in many areas of Iberian spoken language research.
The purpose of this Special Issue is to present recent progress and significant advances in all areas of speech science and technology research in the context of IL. Submitted papers must address topics specific to IL and/or issues raised by analyses of spoken data that shed light on speech science and linguistic theories regarding these languages. Research which deals with IL data, but makes use of standard techniques should not be submitted for this Special Issue. However, both research presenting relevant optimization of current technology and systems, and work exploring specific features of IL spoken corpora will be considered for submission.
This Special Issue is one of the first initiatives proposed by the recently created SIG-IL (ISCA Special Interest Group on Iberian Languages, URL http://www.il-sig.org). The purposes of the SIG-IL are to promote research activities on IL, to sponsor and/or organise meetings, workshops and other events on related topics, and to make speech corpora publicly available by promoting joint evaluation efforts. Furthermore, the SIG-IL is also strongly committed to encouraging world-class research within its community in order to contribute with new ideas to the field of speech science and technology. Original, previously unpublished submissions for the following areas, involving IL and detailing the language-specific aspects, are encouraged:
Topics
o Linguistics, Phonology and Phonetics
o Prosody
o Paralinguistic & Nonlinguistic Information in Speech
o Discourse & Dialogue
o Speech Production
o Speech Perception
o Physiology & Pathology
o Spoken Language Acquisition, Development and Learning
o Spoken Language Generation & Synthesis
o Language/Dialect Identification
o Speech and Speaker Recognition: acoustic, language and pronunciation modeling.
o Spoken Language Understanding
o Multi-modal / Multi-lingual Processing
o Spoken Language Extraction/Retrieval
o Spoken Language Translation
o Spoken/Multi-modal Dialogue Systems
o Spoken Language Resources and Annotation
o Evaluation and Standardization
o Spoken Language Technology for the Aged and Disabled (e-inclusion)
o Spoken Language Technology for Education (e-learning)
o Interdisciplinary Topics in Speech and Language
o New Applications
Guest Editors
Isabel Trancoso INESC-ID, Portugal
Nestor Becerra-Yoma Univ. de Chile, Chile
Plinio A. Barbosa Univ. of Campinas, Brazil
Rubén San-Segundo UPM, Spain
Kuldip Plaiwal Griffith University, Australia
Important Dates
Submission deadline: May 31st, 2007
Notification of acceptance: October 31st, 2007
Final manuscript due: December 30th, 2007
Tentative publication date: March, 2008
Submission Procedure
Prospective authors should follow the regular guidelines of the Speech Communication Journal for electronic submission (http://ees.elsevier.com/specom). During submission authors must select the Section “Special Issue Paper”, not “Regular Paper”, and the title of the special issue should be referenced in the “Comments” (Special Issue on Iberian Languages) page along with any other information.

Papers accepted for FUTURE PUBLICATION in Speech Communication

Full text available on http://www.sciencedirect.com/ for Speech Communication subscribers and subscribing institutions. Free access for all to the titles and abstracts of all volumes and even by clicking on Articles in press and then Selected papers.

top

FUTURE CONFERENCES

Publication policy: Hereunder, you will find very short announcements of future events. The full call for participation can be accessed on the conference websites
See also our Web pages (http://www.isca-speech.org/) on conferences and workshops.

FUTURE INTERSPEECH CONFERENCES

INTERSPEECH 2007-EUROSPEECH
August 27-31,2007,Antwerp, Belgium
Chair: Dirk van Compernolle, K.U.Leuven and Lou Boves, K.U.Nijmegen
Website
INTERSPEECH 2007 is the eighth conference in the annual series of INTERSPEECH events and also the tenth biennial EUROSPEECH conference. The conference is jointly organized by scientists from the Netherlands and Belgium, and will be held in Antwerp, Belgium, August 27-31, 2007, under the sponsorship of the International Speech Communication Association (ISCA).
The INTERSPEECH meetings are considered to be the top international conferences in spoken language processing, with more than 1000 attendees from universities, industry, and government agencies. The conference offers the prospect of meeting the future leaders of our field, exchanging ideas, and exploring opportunities for collaboration, employment, and sales through keynote talks, tutorials, technical sessions, exhibits, and poster sessions.
In recent years the INTERSPEECH meetings have taken place in a number of exciting venues including most recently Pittsburgh, Lisbon, Jeju Island (Korea), Geneva, Denver, Aalborg (Denmark), and Beijing.
CALL FOR PAPERS
AREAS AND TOPICS OF INTEREST:
Interspeech is the world's largest and most comprehensive conference on Speech Science and Speech Technology and it solicits papers in the following areas and topics:
A.Human speech production, perception and communication
Phonology and phonetics
Discourse and dialogue
Prosody (production, perception, prosodic structure)
Paralinguistic and nonlinguistic cues (e.g. emotion and expression)
Speech production
Speech perception
Physiology and pathology
Spoken language acquisition, development and learning
B.Speech and Language technology
Speech and audio processing
Speech enhancement
Speech coding and transmission
Spoken language generation and synthesis
Speech recognition
Spoken language understanding
Accent and language identification
Cross-lingual and multi-lingual processing
Multimodal/multimedia signal processing
Speaker characterization and recognition
C.Spoken language systems and applications
Dialogue systems
Systems for information retrieval
Systems for translation
Applications for aged and handicapped persons
Applications for learning and education
Other applications
D.Resources, standardization and evaluation
Spoken language resources and annotation
Evaluation and standardization
PAPER SUBMISSION
Prospective authors are invited to submit full-length, four-page papers (including figures and references) via the conference website by March 23, 2007 midnight at the International Date Line West. Since the conference takes place early this year, the paper submission deadline will not be postponed by any significant amount of time.
The paper preparation guidelines can be found on the conference website. They stipulate that authors may submit multimedia files to illustrate their contribution. These multimedia files will have to be incorporated in one ZIP archive.
The conference will host several Special Sessions (see conference website). If authors want their paper to be considered for one of these Sessions, they can specify that during paper submission. Irrespective of this specification, all papers will follow the same electronic review procedure.
Authors will have to declare that their contribution is original and not being submitted for publication elsewhere (e.g., another conference, workshop, or journal).
Each corresponding author will be notified by e-mail of the acceptance or rejection of his paper by May 25, 2007. Minor updates of accepted papers will be allowed during May 25 - June 3, 2007.
More information is available on the conference website
IMPORTANT DATES
Full paper submission deadline: March 23, 2007
Notification of paper acceptance/rejection May 25, 2007
Early registration deadline: June 22, 2007
Further information via website or email.
ORGANIZERS
Professor Dirk Van Compernolle (General Chair)
Professor Lou Boves (General Co-Chair)
c/o Annitta De Messemaeker
Katholieke Universiteit Leuven
Department of Electrical Engineering
Kasteelpark Arenberg 10
B3001 Heverlee
Belgium
Fax: +32 16 321723
Email
Website

INTERSPEECH 2008-ICSLP
September 22-26, 2008, Brisbane, Queensland, Australia
Chairman: Denis Burnham, MARCS, University of West Sydney.

INTERSPEECH 2009-EUROSPEECH
Brighton, UK,
Chairman: Prof. Roger Moore, University of Sheffield.

top

FUTURE ISCA TUTORIAL AND RESEARCH WORKSHOP (ITRW)

AVSP 2007

International Conference on Auditory-Visual Speech Processing 2007,
August 31 - September 3, 2007
Kasteel Groenendael, Hilvarenbeek, The Netherlands
The next International Conference on Auditory-Visual Speech Processing (AVSP 2007) will be organised by different members of Tilburg University (The Netherlands). It will take place in Kasteel Groenendael in Hilvarenbeek (The Netherlands) from August 31, 2007 till September 3, 2007, immediately following Interspeech 2007 in Antwerp (Belgium). Hilvarenbeek is located at close distance from Antwerp, so that attendance at AVSP 2007 can easily be combined with participation in Interspeech 2007.
Auditory-visual speech production and perception by human and machine is an interdisciplinary and cross-linguistic field which has attracted speech scientists, cognitive psychologists, phoneticians, computational engineers, and researchers in language learning studies. Since the inaugural workshop in Bonas in 1995, Auditory-Visual Speech Processing workshops have been organised on a regular basis (see an overview at the avisa website). In line with previous meetings, this conference will consist of a mixture of regular presentations (both posters and oral), and lectures by invited speakers. All presentations will be plenary.
We are happy to announce that the following experts have agreed to give a keynote lecture at our conference: Sotaro Kita (Birmingham)
Asif Ghazanfar (Princeton)
More details about the conference can be found on the website
Further information

Third ITRW on NON-LINEAR SPEECH PROCESSING (NOLISP'07)

May 22-25, 2007 , Paris, France
Website
Many specifics of the speech signal are not well addressed by the conventional models currently used in the field of speech processing. The purpose of the workshop is to present and discuss novel ideas, work and results related to alternative techniques for speech processing, which depart from mainstream approaches.
SUBMISSION
Prospective authors are invited to submit a 3 to 4-page paper proposal in English, which will be evaluated by the Scientific Committee. Final papers will be due 1 month after the workshop, for inclusion in the CD-ROM proceedings. A special issue in Speech Communication (Elsevier) will follows.
KEY DATES
Submission (full paper): 15 January 2007
Notification of acceptance: 23 February 2007
Workshop: 22-25 May 2007
Final (revised) paper: 25 June

6th ISCA Speech Synthesis Research Workshop (SSW-6)

University of Bonn (Germany), August 22-24, 2007
A satellite of INTERSPEECH 2007 (Antwerp)in collaboration with SynSIG and IfK (University of Bonn)
Organized shortly after the 16th International Congress on Phonetic Sciences (Saarbrücken, Germany, August 6-10, 2007). Like its predecessors in Autrans (France) 1990, New Paltz (NY, USA) 1994, Jenolan (Australia) 1998, Pitlochry (UK) 2001, and Pittsburgh (PA, USA) 2004, SSW-6 will cover all aspects of speech synthesis and adjacent fields, such as:
TOPICS (updated list)
* Text processing for speech synthesis
* Prosody Generation for speech synthesis
* Speech modeling for speech synthesis applications
* Signal processing for speech synthesis
* Concatenative speech synthesis (diphones, polyphones, unit selection)
* Articulatory synthesis
* Statistical parametric speech synthesis
* Voice transformation/conversion/adaptation for speech synthesis
* Expressive speech synthesis
* Multilingual and/or multimodal speech synthesis
* Text-to-speech and content-to-speech
* Singing speech synthesis
* Systems and applications involving speech synthesis
* Techniques for assessing synthetic speech quality
* Language resources for speech synthesis
* Aids for the handicapped involving speech synthesis.
Deadlines (updated)
* Full-paper submission (up to 6 pages) - May 1, 2007
* Notification of acceptance - June 25, 2007
* Deadline for paper modification - July 15, 2007
Please send your papers, preferably as PDF files, as an e-mail attachment. Further information can soon be obtained from the website of the workshop,
Contact
Prof. Wolfgang Hess

8th Workshop on Discourse and Dialogue (SIGdial), Antwerp, Belgium

Antwerp, September 2-3, 2007
Held immediately following Interspeech 2007
Continuing with a series of successful workshops in Sydney, Lisbon, Boston, Sapporo, Philadelphia, Aalborg, and Hong Kong, this workshop spans the ACL and ISCA SIGdial interest area of discourse and dialogue. This series provides a regular forum for the presentation of research in this area to both the larger SIGdial community as well as researchers outside this community. The workshop is organized by SIGdial, which is sponsored jointly by ACL and ISCA.
Topics of Interest
We welcome formal, corpus-based, implementation or
analytical work on discourse and dialogue including but not restricted to the following three themes:
1. Discourse Processing and Dialogue Systems
Discourse semantic and pragmatic issues in NLP applications such as text summarization, question answering, information retrieval including topics like:
· Discourse structure, temporal structure, information structure
· Discourse markers, cues and particles and their use
. (Co-)Reference and anaphora resolution, metonymy and bridging resolution
· Subjectivity, opinions and semantic orientation
Spoken, multi-modal, and text/web based dialogue systems including topics such as:
· Dialogue management models;
· Speech and gesture, text and graphics integration;
· Strategies for preventing, detecting or handling miscommunication (repair and correction types, clarification and under-specificity, grounding and feedback strategies);
· Utilizing prosodic information for understanding and for disambiguation;
2. Corpora, Tools and Methodology
Corpus-based work on discourse and spoken, text-based and multi-modal dialogue including its support, in particular:
· Annotation tools and coding schemes;
· Data resources for discourse and dialogue studies;
· Corpus-based techniques and analysis (including machine learning);
· Evaluation of systems and components, including methodology, metrics and case studies;
The pragmatics and/or semantics of discourse and dialogue (i.e. beyond a single sentence) including the following issues:
· The semantics/pragmatics of dialogue acts (including those which are less studied in the semantics/pragmatics framework);
· Models of discourse/dialogue structure and their relation to referential and relational structure;
· Prosody in discourse and dialogue;
· Models of presupposition and accommodation; operational models of conversational implicature.
Submissions
The program committee welcomes the submission of long papers for full plenary presentation as well as short papers and demonstrations. Short papers and demo descriptions will be featured in short plenary presentations, followed by posters and demonstrations.
· Long papers must be no longer than 8 pages, including title, examples, references, etc. In addition to this, two additional pages are allowed as an appendix which may include extended example discourses or dialogues, algorithms, graphical representations, etc.
· Short papers and demo descriptions should aim to be 4 pages or less (including title, examples, references, etc.). Please use the official ACL style files. Submission/Reviewing will be managed by the START system. Link to follow. Papers that have been or will be submitted to other meetings or publications must provide this information (see submission format). SIGdial 07 cannot accept for publication or presentation work that will be (or has been) published elsewhere. Authors are encouraged to make illustrative materials available, on the web or otherwise. For example, excerpts of recorded conversations, recordings of human-computer dialogues, interfaces to working systems, etc.
Important Dates (subject to change)
Submission May 2, 2007
Notification June 13, 2007
Final submissions July 6, 2007
Workshop September 2-3, 2007
Websites
Workshop website:To be announced
Submission website:To be announced
Sigdial website
Interspeech 2007 website
Email
Program Committee (confirmed)
Harry Bunt, Tilburg University, Netherlands (co-chair)
Tim Paek, Microsoft Research, USA (co-chair)
Simon Keizer, Tilburg University, Netherlands (local chair)
Wolfgang Minker, University of Ulm, Germany
David Traum, USC/ICT, USA

CfP-SLaTE Workshop on Speech and Language Technology in Education
ISCA Tutorial and Research Workshop

The Summit Inn, Farmington, Pennsylvania USA October 1-3, 2007.
Website
Speech and natural language processing technologies have evolved from being emerging new technologies to being reliable techniques that can be used in real applications. One worthwhile application is Computer-Assisted Language Learning. This is not only helpful to the end user, the language learner, but also to the researcher who can learn more about the technology from observing its use in a real setting. This workshop will include presentations of both research projects and real applications in the domain of speech and language technology in education.
IMPORTANT DATES
Full paper deadline: May 1, 2007.
Notification of acceptance: July 1, 2007.
Early registration deadline: August 1, 2007.
Preliminary programme available: September 1, 2007.
Workshop will take place: October 1-3, 2007.
LOCATION
The workshop will be held in the beautiful Laurel Highlands. In early October the vegetation in the Highlands puts on a beautiful show of colors and the weather is still not too chilly. The event will take place at the Summit Inn, situated on one of the Laurel Ridges. It is close to the Laurel Caverns where amateur spelunkers can visit the underground caverns. The first night event will be a hayride and dinner at a local winery and the banquet will take place at Frank Lloyd Wright’s wonderful Fallingwater.
TOPICS
The workshop will cover all topics which come under the purlieu of speech and language technology for education. In accordance with the spirit of the ITRWs, the upcoming workshop will focus on research and results, give information on tools and welcome prototype demons
trations of potential future applications. The workshop will focus on research issues, applications, development tools and collaboration. It will be concerned with all topics which fit under the purview of speech and language technology for education. Papers will discuss theories, applications, evaluation, limitations, persistent difficulties, general research tools and techniques. Papers that critically evaluate approaches or processing strategies will be especially welcome, as will prototype demonstrations of real-world applications.
The scope of acceptable topic interests includes but is not limited to:
- Use of speech recognition for CALL
- Use of natural language processing for CALL
- Use of spoken language dialogue for CALL
- Applications using speech and/or natural language processing for CALL
- CALL tutoring systems
- Assessment of CALL tutors

ORGANIZATION-CONTACT
The workshop is being organized by the new ISCA Special Interest Group, SLaTE. The general chair is Dr. Maxine Eskenazi from Carnegie Mellon University .
PROGRAMME
As per the spirit of ITRWs, the format of the workshop will consist of a non-overlapping mixture of oral, poster and demo sessions. Internationally recognized experts from pertinent areas will deliver several keynote lectures on topics of particular interest. All poster sessions will be opened by an oral summary by the session chair. A number of poster sessions will be succeeded by a discussion session focussing on the subject of the session. The aim of this structure is to ensure a lively and valuable workshop for all involved. Furthermore, the organizers would like to encourage researchers and industrialists to bring along their applications, as well as prototype demonstrations and design tools where appropriate. The official language of the workshop is English. This is to help guarantee the highest degree of international accessibility to the workshop. At the opening of the workshop hardcopies and CD-ROM of the abstracts and proceedings will be available.
CALL FOR PAPERS
We seek outstanding technical articles in the vein discussed above. For those who intend to submit papers, the deadline is May 1, 2007. Following preliminary review by the committee, notification will be sent regarding acceptance/rejection. Interested authors should send full 4 page camera-ready papers.
REGISTRATION FEE
The fee for the workshop, including a booklet of Abstracts, the Proceedings on CD-ROM is:
- $325 for ISCA members and
- $225 for ISCA student members with valid identification
Registrations after August 1, 2007 cannot be guaranteed.
ADDITIONAL REGISTRATION INFORMATION
All meals except breakfast for the two and a half days as well as the two special events are included in this price. Hotel accommodations are $119 per night , and breakfast is about $10. Upon request we will furnish bus transport from the Greater Pittsburgh Airport and from Pittsburgh to Farmington at a cost of about $30. ISCA membership is 55 Euros. You must be a member of ISCA to attend this workshop.

ITRW on Evidence-based Voice and Speech Rehabilitation in Head & Neck Oncology

May 2008, Amsterdam, The Netherlands, 2008
Cancer in the head and neck area and its treatment can have debilitating effects on communication. Currently available treatment options such as radiotherapy, surgery, chemo-radiation, or a combination of these can often be curative. However, each of these options affects parts of the vocal tract and/or voice to a more or lesser degree. When the vocal tract or voice no longer functions optimally, this affects communication. For example, radiotherapy can result in poor voice quality, limiting the speaker’s vocal performance (fatigue from speaking, avoidance of certain communicative situations, etc.). Surgical removal of the larynx necessitates an alternative voicing source, which generally results in a poor voice quality, but further affects intelligibility and the prosodic structure of speech. Similarly, a commando procedure (resection involving portions of the mandible / floor of the mouth / mobile tongue) can have a negative effect on speech intelligibility. This 2 day tutorial and research workshop will focus on evidence-based rehabilitation of voice and speech in head and neck oncology. There will be 4 half day sessions, 3 of which will deal with issues concerning total laryngectomy. One session will be devoted to research on rehabilitation of other head and neck cancer sites. The chairpersons of each session will prepare a work document on the specific topic at hand (together with the two keynote lecturers assigned), which will be discussed in a subsequent round table session. After this there will be a 30’ poster session, allowing 9-10 short presentations. Each presentation consists of maximally 4 slides, and is meant to highlight the poster’s key points. Posters will be visited in the subsequent poster visit session. The final work document will refer to all research presently available, discuss its (clinical) relevance, and will attempt to provide directions for future research. The combined work document, keynote lectures and poster abstracts/papers will be published under the auspices of ISCA.
Organizers
Prof. dr. Frans JM Hilgers
Prof. dr. Louis CW Pols, PhD
dr. Maya van Rossum.
Sponsoring institutions:
Institute of Phonetic Sciences - Amsterdam Center for Language and Communication,
The Netherlands Cancer Institute – Antoni van Leeuwenhoek Hospital
Dates and submission details as well as a website address will be announced in a later issue.

top

FORTHCOMING EVENTS SUPPORTED (but not organized) by ISCA

CFP- ETSI Workshop: Speech and Noise in Wideband Communication

22nd & 23rd May 2007, at ETSI Headquarters in Sophia Antipolis, France.
As new types of voice coders, noise cancellation algorithms, transmission technologies and consequently transmission impairments enter the scene and convergence becomes ever more a reality, the standardization community faces new challenges.
Being organised by TC STQ, STF 294 and Mesaqin, under contract to ETSI, the main objectives of the workshop are to:
* Discuss the status, latest advances and trends in wideband speech and audio coding, in particular in the presence of interfering sounds and noise
* Present the results of STF 294: Improving the quality of eEurope wideband speech applications by developing a standardised performance testing and evaluation methodology for background noise transmission
* Exchange information and establish relationships between research, state and industrial organizations involved in the topic
Topics that will be addressed will include speech and audio wideband coding, noise suppression and its artefacts, and quality assessment.
A round table discussion will permit participants to offer views on the current issues and challenges that we will be facing in the future.
Participation in the workshop is free of charge, and open to everyone.
Candidate speakers are invited to send an abstract of their presentation to Jan Holub by Friday 16th March 2007.
For further details, consult the workshop Website For registration please see our web

Séminaire AFCP : Traitement automatique du langage parlé et langues peu dotées

jeudi 21 juin 2007 ; 10h-16h
IMAG / Maison Jean Kuntzmann
Domaine Universitaire de Saint-Martin d'Hères (Grenoble)
accès
Programme préliminaire (liste des interventions) :
-V. Berment (C&S / LIG) "Langues peu dotées : définition et problématiques pour le TALN et le TALP"
-Nimaan Abdillahi, P. Nocera (LIA) "Récents développements du LIA en reconnaissance automatique du Somali"
-Nathalie Vallée (GIPSA / Département Langage et Cognition) : "Organisation syllabique des unités lexicales des langues"
-Thomas Pellegrini, L. Lamel ""Determination d'unites lexicales dans les langues peu-dotees pour la reconnaissance automatique de la parole"
-L. Besacier, V-B Le "Méthodologie du CLIPS pour la reconnaissance automatique de langues peu dotées : application aux langues khmères, vietnamiennes et à l'arabe dialectal"
-Pierette Bouillon (Univ. Geneve, à confirmer) "MedSLT : a multilingual spoken language translation system tailored for medical domains and its deployement for less-resourced languages"
-Dr. Pushpak Bhattacharyya (Department of Computer Science and Engineering Indian Institute of Technology Mumbai, India) : "Spoken Language Technologies for Indian Languages"
Inscriptions : séminaire gratuit mais inscription nécessaire
Ce séminaire est organisé par l'AFCP

ParaLing'07: International workshop on "Paralinguistic speech - between models and data"

Thursday 2 - Friday 3 August 2007
Saarbrücken, Germany
Workshop website
in association with the 16th International Conference on Phonetic Sciences, Saarbrücken, Germany, 6-10 August 2007
Summary of the call for participation
This two-day workshop is concerned with the general area of paralinguistic speech, and will place special emphasis on attempts to narrow the gap between "models" (usually built making strong simplifying assumptions) and "real data" (usually showing a high degree of complexity).
Papers are invited in a broad range of topics related to paralinguistic speech. Papers can be submitted for oral or poster presentation; acceptance for oral presentation is more likely for papers that explicitly address the general theme of the workshop, i.e. "bridging" issues.
There are at least two different versions of bridging: a weak one and a strong one. The weak, more modest one aims at a better mutual understanding, the strong one at profiting from each other's work. We do not know yet whether after these two days, we really will be able to profit from each other in our own work; however, we do hope that we will have reached a level of mutual understanding that will make future co-operation easier.
WORKSHOP THEME
Research on various aspects of paralinguistic and extralinguistic speech has gained considerable importance in recent years. On the one hand, models have been proposed for describing and modifying voice quality and prosody related to factors such as emotional states or personality. Such models often start with high-intensity states (e.g., full-blown emotions) in clean lab speech, and are difficult to generalise to everyday speech. On the other hand, systems have been built to work with moderate states in real-world data, e.g. for the recognition of speaker emotion, age, or gender. Such models often rely on statistical methods, and are not necessarily based on any theoretical models.
While both research traditions are obviously valid and can be justified by their different aims, it seems worth asking whether there is anything they can learn from each other. For example: "Can models become more robust by incorporating methods used for dealing with real-world data?"; "Can recognition rates be improved by including ideas from theoretical models?"; "How would a database need to be structured so that it can be used for both, research on model-based synthesis and research on recognition?" etc.
While the workshop will be open to any kind of research on paralinguistic speech, the workshop structure will support the presentation and creation of cross-links in several ways:
- papers with an explicit contribution to cross-linking issues will stand a higher chance to be accepted as oral papers;
- sessions and proceedings will include space for peer comments and answers from authors;
- poster sessions will be organised around cross-cutting issues rather than traditional research fields, where possible.
We therefore encourage prospective participants to place their research into a wider perspective. This can happen in many ways; as illustrations, we outline two possible approaches.
1. In application-oriented research, such as synthesis or recognition, a guiding principle could be the requirements of the "ideal" application: for example, the recognition of finely graded shades of emotions, for all speakers in all situations; or fully natural-sounding synthesis with freely specifiable expressivity; etc. This perspective is likely to highlight the hard problems of today's state of the art, and a cross-cutting perspective may lead to innovative approaches yielding concrete steps to reduce the distance towards the "ideal".
2. A second illustration of attaining a wider perspective would be to attempt to cross-link work in generative modelling (e.g., expressive speech synthesis) and analysis (e.g., recognition of expressivity from speech). Researchers on generation are invited to investigate the relevance of their work for analysis, and vice versa. What methodologies, corpora or descriptive inventories exist that could be shared between analysis and generation, or at least mapped onto each other? If certain parameters have proven to be relevant in one area, to what degree is it possible to transfer them to the other area? Issues of relevance in this area may include, among other things, personalisation, speaker dependency vs. independency, links between voice conversion in synthesis and speaker calibration in (automatic) recognition or (human) perception, etc.
TOPICS
Paper are invited in all areas related to paralinguistic speech, including, but not limited, to the following topics:
- prosody of paralinguistic speech
- voice quality and paralinguistic speech
- synthesis of paralinguistic speech (model-based, data-driven, ...)
- recognition/classification of paralinguistic properties of speech
- analysis of paralinguistic speech (acoustics, physiology, ...)
- assessment and perception of paralinguistic speech
- typology of paralinguistic speech (emotion, expression, attitude, physical states, ...)
While all papers must be related to paralinguistic speech, papers making the link with a related area, e.g. investigating the interaction of the speech signal with the meaning of the verbal content, are explicitly welcome.
IMPORTANT DATES
1st call for papers 1 December 2006
2nd call for papers 1 February 2007
Deadline for full-paper submission 23April (extended deadline!)
Notification of acceptance 1 June
Final version of accepted papers 15 June
Workshop 2-3 August 2007
LOCATION AND REGISTRATION FEES
The workshop will take place at DFKI on the campus of Saarland University, Germany; on the same campus, the International Conference of Phonetic Sciences will take place during the following week.
Workshop registration fees: To be calculated, but will be around ~150 EUR
SUBMISSIONS
The workshop will consist of oral and poster presentations. Submitted papers will stand a higher chance of being accepted as oral presentations when the relevance to the workshop theme is evident.
Final submissions should be 6 pages long, and must be in English. Word+Latex+OpenOffice templates will be made available on the workshop website.
ORGANISING COMMITTEE
Marc Schröder, DFKI GmbH, Saarbrücken, Germany
Anton Batliner, University of Erlangen-Nürnberg, Germany
Christophe d'Alessandro, LIMSI, Paris, France

CfP-Special session at Interspeech 2007: Novel techniques for the NATO non-native military air traffic controller database (nn-matc)

Following a series of special interest sessions and (satellite) workshops, at Lisbon (1995), Leusden (NL, 1999) and Aalborg (2001), the NATO research task group on speech and language technology, RTO IST031-RTG013, organizes a special session at Interspeech 2007. After having studied various aspects of speech in noise, speech under stress, and non-native speech, the research task group has been studying the effects of all of these factors on various speech technologies.
To this end, the task group has collected a corpus of military Air Traffic Control communication in Belgian air space. This speech material consists predominantly of non-native English speech, under varying noise and channel conditions. The database has been annotated at several levels:
* word transcriptions, which allow research to be conducted on automatic speech recognition and named entity extraction,
* Speaker turns, identified by call signs, allowing for research in speaker recognition and clustering and tracking of conversations.
The database consists of 16 hours of training speech, plus one hour of development and evaluation test sets.
The NATO research task group is making this annotated speech database available for speech researchers, who want to develop novel algorithms for this challenging material. These new algorithms could include noise-robust speaker recognition, robust speaker and accent adaptation for ASR, and context driven named entity detection. In order to facilitate a common task, we have written a suggested test and evaluation plan to guide researchers. At the special session we will discuss research results on this common data set.
More information on the special session, the database and the evaluation plan can be found on the web-site
Submission
Researchers who are interested in contributing to this special session are invited to submit a paper according to the regu