Welcome!
On this Speaker and Speech Recognition website you'll find course materials and other learning resources.
Timetable
Lectures are in TB-01 at 11am-1pm on Tuesdays and 9am-10am on Thursdays.
There are computer-based labs in weeks 2, 3 and 6, which will take place in the Swan/Duck Labs (34/34a-BB-04).
The lecturers for this module are
Dr Philip Jackson
(PJ; yellow) and
Professor Josef Kittler
(JK; orange).
If you have any general queries, you should contact Mrs Amanda Ellis in the Postgraduate Office on AA-02;
otherwise, or if you have an academic question, you can email
Philip Jackson.
Week |
Date |
Lecture/Lab |
Exercise |
|
|
11am-Tu |
4pm-Tu |
9am-Th |
|
1 |
Feb 4 |
Lectures |
|
Lec. |
Ex. 1 |
2 |
Feb 11 |
Lectures |
Lab. 1 |
|
|
3 |
Feb 18 |
Lectures |
Lab. 2 |
|
Ex. 2 |
4 |
Feb 25 |
Reading |
|
5 |
Mar 4 |
Lectures |
|
Lec. |
Ex. 3 |
6 |
Mar 11 |
Lectures |
Lab. 3 |
Lec. |
|
7
| Mar 18 |
Lectures |
|
Lec. |
Ex. 4 |
- |
|
Easter break |
|
8 |
Apr 22 |
Lectures |
|
Lec. |
|
9 |
Apr 29 |
Lectures |
|
Lec. |
Ex. 5 |
10 |
May 6 |
Lectures |
|
Lec. |
|
11 |
May 13 |
Revision |
|
Ex. 6 |
Coursework
The coursework for this module, which is worth 40% of the
credit, comes as three written assignments.
Although each one is associated with an introductory lab session, the assignments must be carried out in your own time, and submitted
before the deadline for each assessment.
The marks
for all three lab assignments are now available (on campus only).
Slides and links
As the weeks go by, the slides that are presented during lectures will be posted below.
There are also
exercises on this web site for you to do in your own time.
You are encouraged to explore the internet yourself for relevant
information.
- Week 1:
- Week 2:
- Week 3:
- Week 5:
- HMM part 2 slides,
eeem034i.pdf
(on campus only)
- Viterbi decoding with an HMM,
applet (on campus only)
- Feature extraction 1 slides,
eeem034h.pdf
(on campus only)
- Week 6:
- Week 7:
- Week 8:
- Week 9:
- Speaker adaptation & noise robustness slides,
eeem034q.pdf
(on campus only)
- Context sensitivity slides,
eeem034o.pdf
(on campus only)
- Week 10:
- Artificial neural networks in ASR slides,
eeem034r.pdf
(on campus only)
Books
The books recommended for this course are:
- Jelinek: Jelinek, F.,
Statistical Methods for Speech Recognition, MIT Press, 1998 [0-262-10066-5].
£37
- Homes & Holmes: Holmes, J.N. and Holmes, W.J.,
Speech Synthesis and Recognition, Taylor & Francis, 2001 [0-748-40857-6].
£26
There are also many useful articles available on the internet, e.g.:
- Young: Young, S.J.,
"Large Vocabulary Continuous Speech Recognition",
IEEE Signal Processing Magazine 13(5): 45-57, 1996.
gzipped postscript
- Young et al.: Young, S.J. et al.,
The HTK Book, CUED (v3.4), 2006.
website
- Rabiner: Rabiner, L.R.,
A tutorial on HMM and selected applications in speech recognition,
In Proc. IEEE, Vol. 77, No. 2, pp. 257-286, Feb. 1989.
Other books that you might find of interest include:
- Jurafsky & Martin: Jurafsky, D. and Martin, J.H.,
Speech and Language Processing: An Introduction to Natural Language Processing, Computational Linguistics and Speech Recognition, Prentice Hall, 2003
[0-131-22798-X].
£40
- Gold & Morgan:
Gold, B. and Morgan, N.,
Speech and Audio Signal Processing: Processing and Perception of Speech and Music, New York: Wiley, 2000
[0-471-35154-7].
£77
- Gold, Morgan & Ellis:
Gold, B., Morgan, N., and Ellis, D.,
Speech and Audio Signal Processing: Processing and Perception of Speech and Music (hardback), 2nd ed., New York: Wiley, 2011
[0-470-19536-3].
£65
|