Subject: ISCApad #35 Date: Thu, 5 Apr 2001 09:19:28 +0100 From: Isabel Trancoso To: isca_members@isca-speech.org ============================================================================ ISCApad number 35 April 5th, 2001 ============================================================================ Dear ISCA members, Welcome to ISCApad #35. Here's the table of contents of this issue: ISCA news: - Message from the Secretariat: Members who need a membership receipt for tax purposes, for this year or previous years, can ask the secretariat at 'info@isca-speech.org'. They will be sent by regular mail. Members who, for any reason, did not receive previous issues of ISCApad can also contact the secretariat at 'info@isca-speech.org'. The last issues will then be sent to them by email individually. News from SIGs: - InSTIL: An Illustrated History of Speech Technology in Language Learning Websites: - As a follow up of the last ISCA workshop on Speech and Emotions in Newcastle, a new website was launched to collect tools, corpora, publications and other relevant information. The address is www.interactivesys.com/emotions Journals: - Special issue of IEEE Transactions on Speech and Audio Processing on Speech Technologies for Mobile and Portable Devices - Special issue of Speech Communication on Speech Processing for Hearing Aids - Special Issue of The Journal of VLSI Signal Processing on Real World Speech Processing Future events: - ISCA Tutorial and Research Workshop on Multi-Modal Dialogue in Mobile Environments Kloster Irsee, Germany, 17-22 June 2002 (Preliminary announcement) - Eurospeech Special Event on Noise Robust Recognition Aalborg, Denmark, September 4th, 2001 http://eurospeech2001.org/ese/NoiseRobust/index.html - CRAC Consistent & Reliable Acoustic Cues for sound analysis One-day workshop directly before Eurospeech-2001 Aalborg, Denmark, Sunday September 2nd, 2001 http://www.ee.columbia.edu/crac/ - Workshop on Evaluation for Language and Dialogue Systems ACL/EACL 2001 Toulouse, France, July 6-7, 2001 http://www.limsi.fr/TLP/CLASS/eacl01.html ISCA Greetings, Isabel Trancoso ============================================================================= AN ILLUSTRATED HISTORY OF SPEECH TECHNOLOGY IN LANGUAGE LEARNING The InSTIL SIG of CALICO, EUROCALL and ISCA has undertaken to complete as soon as possible this year AN ILLUSTRATED HISTORY OF SPEECH TECHNOLOGY IN LANGUAGE LEARNING in the form of an exhibition of 20 large posters to be shown at EUROCALL 2001, Nijmegen, Holland and EUROSPEECH 2001, Aalborg, Denmark and on a web site at http://www.language-speech.com/ (cloaked redirected also from langue-parole.com). For more info. on the InSTIL SIG, see http://dbs.tay.ac.uk/instil/. The Exhibition will also be shown in the USA at InSTIL 2002 in Mid-March, San Diego, California. A model for what we are doing can be found on the web site for the History of Computer Assisted Language Learning at http://www.history-of-call.org/ (this project will also be much enhanced this year). CALL FOR INFORMATION: If you have made any form of contribution relevant to History of the Integration of Speech Technology in Language Learning, please give us a summary with dates, names of people, pictures, etc... If you know someone who has, please tell us so we can contact them, we have already done much research but are grateful for info. from any source. We will naturally link up to any organisation of importance in the field including HLT portals, ELSNET, etc... our parent associations, the EduSIG, other SIGs of ISCA, etc... but if you represent an association of some importance to a sub area of our field, please let us know. This initiative is also linked to our gathering of info. on Resources relevant to InSTIL (the InSTIL database to be published in 2002). THE EXHIBITION: If we have a good enough response, we will include a small number of artifacts to illustrate this history. If you are in a position to lend such an artifact, please let us know. We can give you plenty of examples of hardware, software, courseware or publications artifacts you could contribute. We would reimburse any expenses you incur. INDUSTRY SPONSORS: We were very fortunate last year to enlist the assistance of quite a few industry sponsors (sometimes just in kind), if you are involved in a field which impacts on the Integration of Speech Technology in LL and would like to get involved, please contact Philippe at P.DELCLOQUE@MAIL1.TAY.AC.UK Best regards and thanking you in anticipation, Philippe InSTIL SIG European co-co-ordinator. ============================================================================= CALL FOR PAPERS A Special Issue of IEEE Transactions on Speech and Audio Processing Speech Technologies for Mobile and Portable Devices In the fifty-six years since the inauguration of ENIAC, computer systems have shrunk by a factor of 1,000,000 in power consumption, 100,000,000 in size, and 300,000,000 in weight. Yet in this continuing reduction of computer technology to digital dust, one important dimension remains constant: the size of the human fingertip. Our cellular phones, pagers and PDAs may spread information and the power to analyze it everywhere, but the traditional mouse-and-keyboard mode of human-computer interaction cannot make the leap to handheld devices. Computer speech technology will fill the breach. What could be simpler than pressing a button on a PDA, and posing the question "Am I free for lunch on Thursday?" In sufficiently restricted domains, it is already possible to interact with computers by talking to them. The keyboard and mouse are just so, well, Twentieth Century. Proposed Content We seek outstanding technical articles on speech technology ---recognition, synthesis and understanding--- as implemented on or in service of portable and mobile devices. By these we mean cellular phones, personal digital assistants, automobiles, wearable computers, and other non-desktop devices. We intend to address all pertinent areas of speech processing: analog and digital signal processing, noise rejection, acoustic and language modeling, search algorithms, physical, electronic and software design, low-cost and high-end implementations, on-board, deferred or in-network recognition, and scientific and engineering issues from fundamentals to complete applications. Specific topics for coverage include - engineering and economics of power-, weight- and size-constrained speech processing systems - algorithms for recognition and synthesis in adverse acoustic environments - hardware assistance for speech recognition - characteristics of successful applications - standards for speech coding for remote recognition - case studies in engineering complete products (speech-enabled cellular phone, PDA or car) However, this list is not intended to be exhaustive, and any high quality technical discussion of an appropriate subject will be entertained for publication. Submission Schedule Submission deadline: September 1, 2001 Notification: October 30, 2001 Publication date: May 2002 How to Submit Interested authors should send five (5) copies of a hardcopy manuscript, or a Postscript file as an email attachment, to either of: Dr. Harry Printz Prof. Isabel Trancoso IBM Watson Research Center INESC PO Box 218 R. Alves Redol, 9 Yorktown Heights, NY 10598 1000 Lisbon USA PORTUGAL printz@us.ibm.com Isabel.Trancoso@inesc.pt ============================================================================= Speech Communication Call For Papers Special Issue on Speech Processing for Hearing Aids Guest Editors: Fa-Long Luo, Bernard Widrow and Caslav Pavlovic With the widespread usage of digital hearing aids and increasing demand on their performance, speech processing techniques are playing a more important role in modern hearing aid systems. Speech processing for hearing aids includes several topics such as acquisition, transfer, amplification, transmission, measurement, filtering, parameter estimation, separation, detection, enhancement and classification of speech signals. Because of the limitations imposed by the hardware requirements, computational speed, power supply and other practical factors, the development and implementation of speech processing techniques for hearing aids remains a challenging and active area of research. This special issue aims to provide a high quality forum for scientists and engineers interested in hearing-aid design in both the academic and industrial communities to help guide in the development of new and improved hearing aids. In addition, it is hoped that the issue will attract a broad audience in the speech processing community. Topics of interest include but are not limited to: (1) DSP architectures, complexity and parallel implementation in hearing aids. (2) Speech enhancement and noise reduction. Pause and noise detection. Advanced spectral subtraction methods. Source separation. Signal and noise representations and modeling. Listening environment classification. (3) Multichannel and array signal processing. Beamforming, directionality. Binaural listening. (4) Perceptually-based processing algorithms. Characterization and compensation of hearing loss. Spectral enhancement and frequency transposition. Filter bank, multi-rate, multi-band, frequency and transform domain processing for hearing aids. Sampling, extrapolation and interpolation. (5) Feedback cancellation. Reduction of device internal noise and artifacts. (6) System tests, performance evaluation, calibration and parameter fittings. (7) Adaptive signal processing, nonlinear signal processing, higher-order based signal processing, intelligent signal processing for hearing aids. The schedule of the special issue is as follows. Manuscript Submission: July 1, 2001 Notification of Decision: September 30, 2001 Final Manuscripts due: December 16, 2001 Tentative Publication date: April, 2002. Prospective authors should follow the regular paper preparation guidelines of Speech Communications (http://www.elsevier.com/inca/publications/store/5/0/5/5/9/7/) and send six hard copies or electronic file ( .pdf or .ps format) to: Dr. Fa-Long Luo GN ReSound Corporation 220 Saginaw Drive Redwood City, CA 94063 USA Fax: 650-261-2284 email:fluo@gnresound.com ============================================================================= Call For Papers Special Issue on Real World Speech Processing The Journal of VLSI Signal Processing To welcome the advent of the 21st century, the Journal of VLSI Signal Processing is planning to publish a special issue on Real World Speech Processing. Speech processing has made tremendous progress in the last 20 years of the 20th century. Many practical applications have been demonstrated and new research directions have been identified. However, making speech technologies useful in the real world, particularly as a natural interface to intelligent services, is still a challenge. Compared to a human listener, for example, most of the spoken language systems do not perform well when actual operating conditions deviate from the intended ones. This special issue of the Journal of VLSI Signal Processing will focus on the most recent research and development aiming at making speech technologies applicable to the real world. It will address a wide range of speech technologies including speech analysis, recognition, understanding, text-to- speech, and speech coding. We solicit submission of papers in the following technical areas: - Speech analysis - spectral analysis, pitch extraction, and other speech features - Robust speech recognition - robust speech modeling, model compensation and adaptation, normalization techniques for noise, channel and speaker effects - Speech understanding - robust language modeling, parsing, semantic processing & classification - Linguistic detection - keyword spotting, utterance verification and rejection - Robust speaker verification and identification - Distant-talking speech signal acquisition and enhancement - Echo cancellation for hands-free speech communication - Microphone and microphone array for hands-free speech communication - Multi-modal voice-enabled communication interface - Speech synthesis for noisy environments - Acoustic environment database and tools - Robust speech coding for noisy and impaired channels Submission of manuscripts and important dates: Sep. 30, 2001: Due date for 5 copies of full manuscripts Feb. 28, 2002: Notification of acceptance/rejection Apr. 15, 2002: Due date for final version Third/fourth quarter, 2002: Scheduled publication date Submissions should be made direct to the Guest Editor/Editor Guest Editors Prof. Sadaoki Furui Dept. of Computer Science Tokyo Institute of Technology 2-12-1 Ookayama Meguro-ku Tokyo 152-8552 Japan Phone/ Fax: 3-57343480 email: furui@cs.titech.ac.jp Dr. Biing-Hwang Juang Bell Labs, Lucent Technologies 600 Mountain Avenue, 2D-535 Murray Hill NJ 07974-0636 Phone: 908-582-2002 email: bjuang@lucent.com Editor Prof. Jhing-Fa Wang Dept. of Electrical Eng. National Cheng Kung University Tainan, Taiwan, ROC Phone: 886-6-2746867 Fax: 886-6-8746867 email: wangjf@server2.iie.ncku.edu.tw ============================================================================= Preliminary Announcement: After the successful workshops initiated by Paul Dalsgaard - IDS-95: Spoken Dialogue Systems - Theories and Applications Vigso, 1995 - IDS-99: Interactive dialogue in multi-modal systems; Kloster Irsee, Germany, 1999 we are pleased to announce ISCA Tutorial and Research Workshop MULTI-MODAL DIALOGUE IN MOBILE ENVIRONMENTS Kloster Irsee, Germany 17-22 June 2002 Speech and display-oriented dialogue and their integration for pedestrian and in-car use Bringing together academic & industrial research with developers & practitioners from industry Organization Committee: Elisabeth André (Univ. Augsburg, Germany), Laila Dybkjaer (Univ. of. Southern Jutland, Odense, Denmark), Paul Heisterkamp & Wolfgang Minker (DaimlerChrysler Research, Ulm, Germany) Further information soon to be announced: {Paul.Heisterkamp|Wolfgang.Minker}@DaimlerChrsyler.com The workshop will cover all aspects of spoken, but also multimodal interaction in mobile environments and will emphasize design issues and applications. Moreover, the workshop will focus on research and results, give information on tools and run prototype demonstrators of the expected future application. The format of the workshop will be a non-overlapping mixture of oral and poster sessions. A number of tutorial lectures will be given by internationally recognised experts from the area of Spoken Dialogue Systems. 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. It is our belief that this general format will ensure a lively and valuable workshop. The organisers would like to encourage researchers and industrialists to take the opportunity to bring their applications as well as their demonstrator prototypes and design tools for demonstration to the workshop. If sufficient interest is shown, a special demonstrator/poster session will be organised and followed by a discussion session. The Proceedings from the workshop will be available at the opening of the workshop. The official language of the workshop and the Proceedings is English. An edition of Lecture Notes in Computer Science collecting full-paper versions of excellent papers presented at the workshop is planned. The workshop focus on research issues, applications and tools concerned with the following interdisciplinary topics which are all part of general research and development of the main areas of Spoken Dialogue in Mobile Environments. Papers may discuss theories, applications, evaluation, limitations, general tools and techniques. Discussion papers that critically evaluate approaches or processing strategies and prototype demonstrations are especially welcome. Speech and its Interpretation robust algorithms for automatic speech recognition and understanding in mobile environments spontaneous, conversational and disfluent speech spoken dialogue processing speech and language modelling microphone and processing devices Design Issues dialogue control situation awareness pro-active dialogue geo-position awareness spatial reference (input and output) attention raising multimodality an safety considerations Applications applications for mobile environments coordination of multiple applications conflict description and solving Dialogue Systems Architecture integration of local and server-based dialogue technical integration management Evaluation of Dialogue Strategies and Systems strategies and paradigms especially adapted to mobile environments ============================================================================= Eurospeech Special Event on Noise Robust Recognition: Invitation to submit a paper. As you may already be aware, a special event on Noise Robust Recognition is being organized as part of Eurospeech 2001 (4th Sept 2001). What makes this event special is that all papers have to present results on a common database so we can all learn more about the relative performance of the different techniques. On behalf of the organizing committee we'd like to invite you to submit a paper for this event. The greater coverage of different algorithms represented, the more we can all learn. You'll be pleased to hear that the deadline for submission of papers has been extended from the 30th March to 13 May 2001, so there is still time! (If you don't already have a copy of the Aurora 2 database, you'll have to be quick in placing your order with ELRA!) We already have promises of 12 papers so the success of the event is assured. It will be even more valuable in giving a snapshot of the state-of-the-art of noise robust algorithms (and their relative performance) if further researchers are able to participate with papers. This motivation and knowing that some groups would appreciate more time to process the Aurora 2 databases, encouraged us to extend the deadline. The new deadline still leaves time for paper review and publication in the Eurospeech proceedings. We have been approached by a publisher interested in making a book based on papers at the event. It would be great to have your work included too in this publication of the state-of-the-art. Full information on the ESE is available from the Eurospeech web site The highlights of recent changes are: 1) Revised schedule May 13, 2001: Submission of full, four-pages, paper for publication (results on Aurora 2 must be included). This paper will appear in the Eurospeech 2001 Proceedings. June 8, 2001: Notification of acceptance 2) Web only papers Authors with papers accepted for publication in the Proceedings may also provide an updated paper that will be accessible at the Eurospeech website. There is no 4 page limit on this paper and authors may like to use this for: a) extended papers providing more comprehensive results, b) updates to results ahead of the conference showing the latest algorithm improvements and results. The deadline for submission of these extended papers is August 3. August 3, 2001: Deadline for Web-only paper (containing most recent algorithm improvements and results). There is no four-page limit on this paper. Papers will appear at 3) Book The organisers are in contact with a publisher with the purpose of making a peer-reviewed book presentation of some, or all, of the papers. 4) Spreadsheet and format for presentation of results These are now available for download from the ESE web site. We look forward to seeing your paper and to a successful and enjoyable event in September. Best Regards David Pearce Chair ESE on Noise Robust Recognition ============================================================================= REMINDER: CRAC Workshop on Consistent & Reliable Acoustic Cues Aalborg, Sunday Sep 2nd 2001, immediately before Eurospeech. http://www.ee.columbia.edu/crac/ * What's it about? The CRAC workshop on Consistent and Reliable Acoustic Cues for Sound Analysis will bring together researchers in - Sound organization and computational auditory scene analysis - Robust speech recognition and missing data techniques - Other aspects of human and computer perception of real-world audio * What will it be like? Participation will be limited to about 40 attendees to foster a true workshop atmosphere. There will be lecture, poster and discussion sessions in the morning and afternoon of the one-day workshop. * How can I participate? Submit a 500 word abstract by April 30th 2001 to: crac-abstract@ee.columbia.edu * How do I decide between submitting my work to Eurospeech or to CRAC? - CRAC will be a focussed forum dealing with emerging techniques in sound analysis. This will allow a different emphasis in your paper, and a higher level of discussion at the workshop. - We will be editing a special issue of Speech Communication containing expanded versions of papers from this workshop. - The papers submitted to the workshop will be made permanently available via the internet on the conference web site. - The deadline for a 500 word abstract is April 30th, and the full 4-page papers don't have to be finished until July 31st. * Where can I get more information? Visit the CRAC web site, http://www.ee.columbia.edu/crac/ ============================================================================= Call for Papers Workshop on Evaluation for Language and Dialogue Systems ACL/EACL 2001 Toulouse, France July 6-7, 2001 WORKSHOP GOALS The aim of this two day workshop is to identify and to synthesize current needs for language-technology evaluation. The first day of the workshop will focus on one of the most challenging current issues in language engineering: the evaluation of dialogue systems and models. The second day will extend the discussion to address the problem of evaluation in language engineering more broadly and on more theoretical grounds. The space of possible dialogues is enormous, even for limited domains like travel information servers. The generalization of evaluation methodologies across different application domains and languages is an open problem. Review of published evaluations of dialogue models and systems suggests that usability techniques are the standard method. Dialogue-based system are often evaluated in terms of standard, objective usability metrics, such as task-completion time and number of user actions. In the past, researchers have proposed and debated theory-based methods for modifying and testing the underlying dialogue model, but the most widely used method of evaluation is usability testing, although more precise and empirical methods for evaluating the effectiveness of dialogue models have been proposed. For task-based interaction, typical measures of effectiveness are time-to-completion and task outcome, but the evaluation should focus on user satisfaction rather than on arbitrary effectiveness measurements.Indeed, the problems faced in current approaches to measurement of effectiveness dialogue models and systems include: Direct measures are unhelpful because efficient performance on the nominal task may not represent the most effective interaction Indirect measures usually rely on judgment and are vulnerable to weak relationships between the inputs and outputs Subjective measures are unreliable and domain-specific For its first day, the workshop organizers solicit papers on these issues, with particular emphasis on methods that go beyond usability testing to address the underlying dialogue model. Representative questions to be addressed include: o How do we deal with the combinatorial explosion of dialogue states? o How can satisfaction be measured with respect to underlying dialogue models? o Are there useful direct measures of dialogue properties that do not depend on task efficiency? o What is the role of agent-based simulation in evaluation of dialogue models? Of course, the problems faced in evaluating dialogue and system models are found in other domains of language engineering, even for non-interactive processes such as part-of-speech tagging, parsing, semantic disambiguation, information extration, speech transcription, and audio document indexing. So the issue of evaluation can be viewed at a more generic level, raising fundamental, theoretical questions such as: o What are the interest and benefits of evaluation for language engineering? o Do we really need these specific methodologies, since a form of evaluation sould always be present in any scientific investigation? o If evaluation is needed in language engineering, is it the case for all domains? o What form should it take? Technology evaluation (task-oriented in laboratory environment) or field/user Evaluation (complete systems in real-life conditions)? We have seen before that the the evaluation of dialogue models is still unsolved, but for domains where metrics already exists, are they satisfactory and sufficient? How can we take into account or abstract from the subjective factor introduced by human operators in the process? Do similarity measures and standards offer appropriate answers to this problem? Most of the efforts focus on evaluating process, but what about the issue of language resources evaluation? For its second day of work, the workshop organizers solicit papers on these issues, with the intent to address the problem of evaluation both from a broader perspective (including novel applications domains for evaluation, new metrics for known tasks and resource evaluation) and a more theoretical point of view (including formal theory of evaluation and infrastructural needs of language engineering). NOTE: People who would like to submit a paper on lexical semantic disambiguation evaluation should consider the parallel workshop, on July 5-6, for the closure of the SENSEVAL-2 evaluation campaign. WORKSHOP ORGANIZATION The organization of each of the two days of the workshop will reflect the workshop's two main themes. Each day will begin with a session of presentations of selected papers and follow with panel discussions to synthesize and develop possible methodologies from additional selected workshop papers. WORKSHOP PARTICIPATION The workshop seeks participation from people involved or interested in the problem of evaluation in language processing and the research and industrial communities that study and implement dialogue models for natural-language interaction systems. The first part of the workshop will specifically draw on the natural-language interaction community, for instance like the one developing at the confluence of SIGdial and SIGCHI, which will find in this workshop an atmosphere more flavored by computational-linguistics related issues (see, for example, the First SIGdialWorkshop on Discourse and Dialogue). The second part of the workshop is intended to provide a forum for a broader audience more in the spirit of the one that attended the LREC'2000 Satellite Workshop on Evaluation (see http://www.limsi.fr/TLP/CLASS), in particular offering an opportunity to people involved in language engineering evaluation (e.g ., the CLASS audience) in the context of national or transnational projects or programs, both in Europe and abroad. SUBMISSION DETAILS Paper submissions should follow the two-column format of ACL proceedings and should not exceed eight (8) pages, including references. We strongly recommend the use of ACL LaTeX style files or Microsoft Word Style files tailored for this year's conference. They are available from the ACL-2001 program committee Web site at http://acl2001.dfki.de/style/. Papers should be submitted electronically, as either a LaTeX, Word or PDF file to either: Patrick Paroubek, pap@limsi.fr Karen Ward, kward@cs.utep.edu TIMETABLE OF IMPORTANT DATES Deadline for workshop paper submissions: April 6, 2001 Deadline for notification of workshop paper acceptance: April 27, 2001 Deadline for camera-ready workshop papers: May 16, 2001 Workshop date: July 6-7, 2001 WORKSHOP ORGANIZING COMMITTEE David G. Novick, UTEP Joseph Mariani, Limsi - CNRS Candy Kamm, AT&T Labs Patrick Paroubek, Limsi - CNRS Nils Dahlbdck, Linkvping University Frankie James, NASA Ames Research Center Karen Ward, UTEP, kward@cs.utep.edu SCIENTIFIC COMMITTEE David G. Novick, Joseph Mariani, Candy Kamm, Patrick Paroubek Nils Dahlbdck, Frankie James, Karen Ward, Christian Jacquemin Niels Ole Bernsen, Stephane Chaudiron, Khalid Choukri, Martin Rajman Robert Gaizauskas, Donna Harman, Saod Tazi Tentative: Lynette Hirschman, David Pallett, Carol Peters, Jose Pardo Herman Steeneken, Oliviero Stock, Hans Uszkoreit SPONSORS ACL 2001, CLASS, ELRA, ELSNET We also anticipate co-sponsorship from SIGdial. ADDITIONAL INFORMATION Additional information on the workshop, including accepted papers and the workshop schedule, will be made available as needed at http://www.limsi.fr/TLP/CLASS/eacl01.html ============================================================================= All additional information at the ISCA web-site: http://www.isca-speech.org The ISCA secretariat can be contacted at: info@isca-speech.org Requests concerning membership, Speech Communication and ordering Proceedings should be forwarded to the Secretariat. For message distribution at ISCA list contact: public@isca-speech.org Short messages will be forwarded on a monthly scheme basis to all ISCA members. ============================================================================