|
Courses and Workshops
|
| Title: | Computational approaches to phonology |
| Lecturer(s): | Karin Mueller (Universiteit van Amsterdam) |
| Type: | Introductory Course |
| Section: | Language and Computation |
| Week: | Second | | Time: | 9.00-10.30 (Slot 1) |
| Webpage: | http://staff.science.uva.nl/~kmueller/
|
| Reader: | Course will not have a reader | | Room: | TBA |
Description Every area of speech and language processing requires knowledge about
word structure. The phonological structure of words is of special
interest for speech recognition and speech synthesis systems. A major
problem for these systems are unknown words which do not appear in a
pronunciation dictionary, due to productive word-formation processes
occurring in all natural languages.
The aim of computational phonology is to develop both formal methods and
real systems which are able to predict the phonological structure of
words. The proposed course explores the field by (i) introducing the
concept of "phonological structure of words", and (ii) presenting some
of the current machine-learning methods used for predicting these
structures (e.g. probabilistic phonological grammars, and memory-based
learning techniques).
The course is intended for students with a basic knowledge of
probability theory and computational linguistics and can be regarded as
a natural successor to earlier ESSLLI courses and workshops on
probabilistic NLP.
|
© 2003, 2004 LORIA, Carlos Areces. |
2003-11-20
| |
