Courses persons responsible
Konsta Koppinen

Lecturers
Konsta Koppinen

Implementations

	Period 1	Period 2	Period 3	Period 4	Period 5	Summer
Lecture	-	-	-	-	50 h/per	-

(Timetable for academic year 2006-2007)

Objectives
The objective is to familiarize the student with the practical implementation of a speech recognizer by a project work which carried out independently.

Content

Content	Core content	Complementary knowledge	Specialist knowledge
1.	Implementation of a digit recognizer using HTK (Hidden Markov Model-toolkit).
2.	Calculation of feature vectors, language modelling, estimation of the parameters of an acoustic model using training data, evaluation of a speech recognizer

Requirements for completing the course
Completed project work and a report.

Evaluation criteria for the course

Study material

Type	Name	Auhor	ISBN	URL	Edition, availability...	Exam material	Language
Book	"The HTK Book Version 3.3"	S. Young, G. Evermann, M. Gales, T. Hain, D. Kershaw, G. Moore, J. Odell, D. Ollason, D. Povey, V. Valtchev, P. Woodland			http://htk.eng.cam.ac.uk/docs/docs.shtml	No	English

Prerequisites

Code	Course	Credits	M/R
SGN-4106	SGN-4106 Speech Recognition	5	Recommendable

Prequisite relations (Sign up to TUT Intranet required)

Additional information about prerequisites
The course Speech Recognition can be done at the same time as this course.

Remarks

Replaces the course SGN-4506. Can be done via distance learning in the KIT-network, see the course homepage for instructions.

Scaling

Methods of instruction	Hours
Lectures	2
Laboratory assignments	71
Total sum	73

Correspondence of content
SGN-4506 Speech Recognition Laboratory

Course homepage

Last modified	06.10.2006
Modified by	Konsta Koppinen