POSTDOC POSITION Sound Recognition

[Tjeerd Andringa <tjeerd@xxxxxxxxx>]

POSTDOC POSITION Sound Recognition (1,0 fte, 2 years)

38 hours per week
Auditory Cognition Group (http://www.ai.rug.nl/research/acg/)
University of Groningen (http://www.rug.nl/)

Job description:
This position contributes to the Dutch Companion project, which aimes at
the development of a situational aware robotic assistant for the
Auditory Cognition Group (ACG) of the department of Artificial
Intelligence. ACG focuses on the development of sound recognition
approaches that mimic the versatility of the human auditory system.
Scientifically, the task of the postdoc is to couple explicit top-down
knowledge of sound sources and situations to bottom-up estimated
patterns of signal components. The postdoc is expected to be a strong
team player who is interested in the research of other members and who
can help in management and acquisition of research projects.

We offer a challenging position in a young and ambitious AI-department
of 30+ researchers. We also offer excellent opportunities for defining
your own follow-up projects.

Requirements:
The candidate
    * has a PhD with a focus on modeling natural phenomena in fields
such as physics, artificial intelligence, psychology, or computational
neuroscience.
    * has a broad scientific interest that includes perception and
cognition
    * is able to think abstractly on the architecture level of a
cognitive system and be able to implement ideas to the level of
impressive demo's
    * has an excellent academic record (experience and publications),
is curious, creative and ambitious
    * is fluent in English
    * has affinity with both fundamental research and as engineering
solutions
    * is a team player, open-minded, and actively participates in
knowledge sharing within highly interdisciplinary teams.

Organization:
University of Groningen
Department of Artificial Intelligence.
ALICE (Artificial Intelligence and Cognitive Engineering) is a young,
active, and vibrant, research group in Artificial Intelligence,
comprising of 30+ researchers. Its research is divided into four
interrelated topics: Autonomous Perceptive Systems, Cognitive Modeling,
Multi-Agent Systems, and Language, Sound & Cognition. The department is
responsible for a bachelor and two master programs on AI and Cognitive
Science and has 250 enrolled students.

About the research group:
The recently started research group Auditory Cognition comprises 2
Faculty and 5 PhD-students and is growing. The group investigates the
optimal use of the information conveyed by sounds through the study of
natural auditory systems and their integration with other cognitive
processes. The Auditory Cognition group proves the quality of its
scientific results through the development of intelligent sound
processing systems that can deal with the complexities of an
unconstrained world using models from physics, AI and Cognitive Science.

The Auditory Cognition Group started little more than a year ago on a
basis of not yet disclosed research performed at the University of
Groningen and Sound Intelligence (see
http://www.soundintel.com/bbc1clip-nl.html for an example of a product
in which part of this research is applied). Some of the approach of the
group is published in a thesis
(http://dissertations.ub.rug.nl/faculties/science/2001/t.c.andringa/thesis.pdf)

The approach of the group is typical for the field of Computational
Auditory Scene Analysis (CASA) in which evidence of separate physical
processes is to be assigned to separate representations. The role of
cognition is to help us design systems that are able to deal with the
complexities of unconstrained real-life input. The resulting systems are
not only intended as a cognitively plausible model of audition, but also
as useful engineering systems that are able to function reliably in the
same acoustic environments as humans can.

The postdoc project has two main goals:

    * To contribute to the development of a system, described in figure
2 of Andringa & Niessen (2006). The postdoc must develop an environment
in which bottom-up generated patterns of signal components are matched
to top-down knowledge about the type of patterns that are expected for a
specific interpretation of the input.
    * To make the iCat respond to sound in a natural way. The iCat
ought to orient itself to the direction of important sounds and it ought
to detect novel sounds and respond to these sounds. This part of the
project may involve the application of pragmatic solutions to ensure a
seamless integration with the work of the project partners of Philips,
Decis Lab, and the University of Utrecht.

The position provides excellent opportunities for future (successful)
grant applications.

Additional Information:
Additional information about the vacancy can be obtained from
AcademicTransfer or Dr. Tjeerd Andringa (email: tjeerd@xxxxxxxxx).


Application

You can apply for this job before 10-03-2007 by sending your application to:

University of Groningen
Personnel and Organization Department
P.O. Box 72
9700 AB Groningen
The Netherlands
E-mail address: vmp@xxxxxxxxxxxxx

When applying for this job always mention the vacancy number AT207032.