Course Catalog 2007-2008

SGN-4106 SPEECH RECOGNITION, 5 cr
Speech Recognition

Courses persons responsible
Konsta Koppinen

Lecturers
Anssi Klapuri

Lecturetimes and places
Per V: Monday 12 - 14, TB214
Per V: Wednesday 12 - 14, TB214

Implementations
  Period 1 Period 2 Period 3 Period 4 Period 5 Summer
Lecture - - - - 4 h/week -
Exercise - - - - 4 h/week -
Exam  
(Timetable for academic year 2007-2008)

Objectives
Teaches the basics of automated speech recognition, particularly the use of hidden Markov models and neural networks.

Content
Content Core content Complementary knowledge Specialist knowledge
1. Front-end of a speech recognizer, cepstral coefficients        
2. Training of Hidden Markov models (HMM)       
3. Adaptation methods       
4. Language models       

Requirements for completing the course
Final examination and exercises.

Evaluation criteria for the course

  • Exam and exercises.

  • Used assessment scale is numeric (1-5)

  • Study material
    Type Name Auhor ISBN URL Edition, availability... Exam material Language
    Book "Fundamentals of Speech Recognition" Rabiner, L. & Juang, B-H.     Prentice Hall, 1993 No  English 
    Book "Pattern Classification" Duda, R.O., Hart, P.E. & Stork, D.G.     Wiley, 2000 No  English 
    Book "Neural Networks - Second Edition" Haykin, S.     Prentice Hall, 1999 No  English 
    Book "Automatic Speech and Speaker Recognition-Advanced Topics" Lee, C.H.,Soong, F.K.,Paliwal,K.K.     Kluwer Academic Publishers, 1996 No  English 
    Book "Robust Speech Recognition in Embedded Systems and PC Applications" Junqua, J.C.     Kluwer Academic Publishers, 2000 No  English 
    Lecture slides "Speech Recognition" Konsta Koppinen       Yes  English 
    Journal "Statistical Pattern Recognition: A Review" Jain, Duin, Mao     IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. 22, No. 1, Jan. 2000 No  English 
    Journal "A tutorial on hidden Markov models and selected applications in speech recognition" Rabiner, L.R.     Proceedings of the IEEE , Volume: 77 Issue: 2 , Feb. 1989 No   

    Prerequisites
    Code Course Credits M/R
    MAT-20500 MAT-20500 Probability Calculus 3 Recommendable
    SGN-2506 SGN-2506 Introduction to Pattern Recognition 4 Mandatory
    SGN-4010 SGN-4010 Speech Processing Methods 2 Mandatory

    Prequisite relations (Sign up to TUT Intranet required)

    Remarks

    Can be done via remote learning in the KIT-network, see the course homepage for instructions.

  • The course is suitable for postgraduate studies.

  • Distance learning

  • ITC utilized during the course

  • - In information distribution via homepage, newsgroups or mailing lists, e.g. current issues, timetables
    - In compiling teaching material, particularly for online use or other electronic media
    - In distributing and/or returning exercise work, material etc

  • Estimate as a percentage of the implementation of the course
  • - Contact teaching: 30 %
    - Distance learning: 0 %
    - Proportion of a student's independent study: 70 %

    Scaling
    Methods of instructionHours
    Lectures 48
    Exercises 72
    Assignments 1

    Other scaledHours
    Preparation for exam 15
    Exam/midterm exam 3
    Total sum 139

    Correspondence of content
    8003163 Speech Recognition

    Course homepage

    Last modified 17.08.2007
    Modified bySari Peltonen