[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[CFP] 3rd IJCAI Workshop on CASA (CASA'99)

Please distribute.

                        Hiroshi "Gitchang" Okuno, Ph.D
                        Kitano Symbiotic Systems Project, ERATO,
                        Japan Science and Technology Corp.
                        6-31-15 Jingumae, Mansion 31 Suite 6A
                        Shibuya-ku, Tokyo 150-0001, Japan.
                        Tel: +81-3-5468-1661, Fax: +81-3-5468-1664
                        E-mail: okuno@nue.org
                        URL:    http://www.symbio.jst.go.jp/~okuno/

                         Call For Papers (CASA'99)

                       The Third IJCAI Workshop on
                 Computational Auditory Scene Analysis
                           Stockholm, SWEDEN
                         Sunday, August 1, 1999

Over the last decade several research communities have begun sustained
research inquiries into the problem of auditory scene analysis, or
the process of decomposing arbitrary sound mixtures, such as those
often found in human listeners' acoustic environments, into component
sounds. Such mixtures typically include non-speech and musical sounds,
the understanding of which represents a challenging area of
psychoacoustic research.

Computational Auditory Scene Analysis (CASA) deals with the problem of
getting software and/or hardware systems not only to interpret sound
mixtures as combinations of their components, but also to decide what
feature set(s) to use as the basis for reconstructing (modelling)
the combination.

Typically, CASA systems rely on or are inspired by psychoacoustic or
psychophysical theories of perception.  However, given that the problem
of auditory scene analysis and its automation is intimately related to
questions studied by other fields ranging from robotics to speech
recognition to machine perception and artificial intelligence, the
CASA community has a strong interdisciplinary component.

The purpose of the IJCAI-99 workshop on Computational Auditory Scene
Analysis (CASA'97) is to bring together researchers from various
disciplines including artificial intelligence, automatic speech
recognition, signal processing, psychoacoustics and psychophysics,
and robotics, and application engineers who are engaged in or interested
in computational auditory scene analysis.  Through key presentations and
ample discussions, it is hoped that the workshop will facilitate the
exchange of ideas among researchers, as well as to bridge the gap
between basic researchers and application engineers.

Having CASA'99 at IJCAI-99 is particularly important as some topics of
the workshop -- including listening to several things simultaneously
or understanding non-speech sounds -- have been also proposed as
Challenge Problems for Artificial Intelligence at the American
Association of Artificial Intelligence's (AAAI) 1996 National Conference.
Indeed, the workshop committee encourages (but does not require!)
potential attendees to consider submitting papers on systems for
interpreting acoustic signals to the parent IJCAI conference.

The list of challenge problems proposed at CASA'95 is available at
the CASA'97 Web page (http://www.nue.org/CASA97/).


The workshop committe is currently soliciting submission of quality
contributions on computational auditory scene analysis.  We strongly
encourage designers of CASA systems to consider giving real-time
demonstrations of their systems.

Topics of interest include but are not limited to the following:

Modeling Issues:
        Cognitive Modeling
        Low-level Auditory Models. etc.

Sound Processing Issues:
        Auditory Filters, Wavelets,
        Time-frequency Distributions,
        Multi-resolution Analysis,
        Adaptation of Signal Processing Parameters, etc.

Representational Issues:
        Auditory/Sound Representation,
        Speech Representation,
        Representation for Music,
        Unified Representation of (possibly
        Dynamic) Vision and Audition, etc.

Architectural Issues:
        Unified Architectures
        Blackboard Architectures
        Multi-Agent Paradigm
        Performance Evaluation
        Hybrid Approach to Top-Down/ Bottom-Up processing, etc.

        Sound Understanding
        Speech Understanding
        Music Understanding
        Performance Evaluation
        Multi-Modal Integration, etc.

The workshop committee anticipates publishing accepted papers
in a workshop proceedings.


Authors will submit a copy of a full paper (limited to 5000 words), or
an extended abstract (approx. 2500 words) electronically to


or by surface mail to

        Frank Klassner
        Computing Sciences Dept.
        Villanova University
        800 Lancaster Avenue
        Villanova, PA 19085

by February 21, 1999.

Electronic submissions are strongly encouraged.  The e-mail should
contain an uuencoded gzipped (or compressed) PostScript file, or
Adobe Acrobat file.
All submitted papers will be reviewed by the workshop committee.


Papers due:                   February 21, 1999
Notification of Acceptance:   March 22, 1999
Camera-ready edition due:     April 20, 1999


Workshop attendees must note that workshop participation is not
possible WITHOUT REGISTRATION for the main conference (International
Joint Conference on Artificial Ingelligence,
web address: http://www.dsv.su.se/ijcai-99/


Frederic Berthommier
Institut de la Communication Parlee, FRANCE

Martin P. Cooke
University of Sheffield, UK

Daniel Ellis
ICSI, Berkeley, USA

Frank Klassner (chair)
Villanova University, USA

Hiroshi G. Okuno
Kitano Symbiotic Systems Project, ERATO, JST, JAPAN

Malcolm Slaney
Interval Research, Inc.,  USA

Email to AUDITORY should now be sent to AUDITORY@lists.mcgill.ca
LISTSERV commands should be sent to listserv@lists.mcgill.ca
Information is available on the WEB at http://www.mcgill.ca/cc/listserv