T-61.182 Neural Networks for Speech Processing
Special course on Information Technology
II (4 ov, L)
Lecturer: Prof. Mikko Kurimo
Assistant: DI Vesa Siivola
Semester: Spring 2002
Credit points: 4 cr (or 2 cr)
Place: Seminar room T4
Time: Thursdays 14-16, starting from January 17
Language: English or Finnish
Seminar course description
Throughout the history of neural computing,
automatic speech processing has been one of its most important and
most demanding application areas.
Speech is also one of the most important means of human communication.
Great efforts have been made to develop speech telecommunication technology
such as telephony.
Therefore, it is not a coincident that the problems of automatical
speech processing lead researchers to study human brains and try to
imitate the information processing methods that could take place there.
The purpose of this course is to provide better understanding of the
speech processing technologies using neural networks.
The emphasize is to give a comprehensive introduction to traditional
and recently developed training methods used in
neural network-based recognizer design attempts.
Important methods presented are, for example, minimum classification error
training and hybrid methods using neural networks and hidden Markov models.
In addition to speech recognition, other speech related topics
such as speaker recognition, voice conversion, speech coding,
and speech enhancement are discussed.
The basic knowledge of neural networks algorithms,
Bayes decision theory, pattern recognition, and speech processing methods
are helpful as prerequisites.
The course is based on:
Handbook of Neural Networks for Speech Processing, edited by Shigeru Katagiri, Artech House (2000).
Requirements for passing the course
To pass the course (4cr), you have to
To pass with distinction, the seminar talk, the handouts, and the
project work must each be very good, and 95% of the exercises
should be solved.
Without the project work the course can be completed to obtain two
- participate actively,
- give a seminar talk on a chapter or part of the book
- generate one or two homework questions or exercises about your
assigned area for other people to solve,
- solve a set of exercises given during the seminar (at least 60%), and
- carry out a small project work, that is, a small-scale research project.
Signing up for the course
Preferably by WWWTopi
or by showing up at the first meeting on January 17th.
If you cannot make it, send e-mail to Mikko.Kurimo@hut.fi.
Relationship to other studies
At TKK the course is suited for the Language Technology major
(Kieliteknologian pää/sivuaine) and for studies in Information
Also students and staff from the KIT
(Kieliteknologian opetuksen verkosto) are welcome -- please sign
up in advance.
Mikko.Kurimo@hut.fi (tel. 451 5388)
Thursday, 17-Jan-2002 15:28:39 EET