Speech Signal ProcessingFall 2003

Tuesdays, 9:10 ~12:00 AMInstructor: Berlin Chen

Topic List and Schedule:

DateTopicHomework / Project9/9

Course Overview & Introduction 9/16

Spoken Language Structure

Homework-1:Depict a spectrogram of a speech utterance with your own name pronounced. (Due: 9/30)

(Please observe the formants and harmonics of the fundamental frequency)

See Results9/23

Hidden Markov Models (I)

Homework-2:Solving the Problems 1* and 2** for HMM (Due: 10/15)

( *Problem 1 should be solved with Forward Algorithm and Backward Algorithm, respectively.

**Problem 2 should be solved with Viterbi Algorithm in both forward and backward directions.)9/30

Hidden Markov Models (II)

Homework-3:Solving the Problem 3 for HMM (Baum-Welch Training) (Due: 10/28)10/7

Hidden Markov Models (III)

- Expectation Maximization (EM) Algorithm

- Review of Estimation Theory10/14

Review of Digital Signal Processing

10/21

Review of Digital Signal Processing

Speech Signal Representations

Project-1:Small-Vocabulary, Isolated Word Recognition (Due 11/10)

10/28

Midterm

11/4

Speech Signal Representations

Linear Prediction Coding of Speech Signals

Project-2:linear prediction coding (Due 11/28)

11/11

Linear Prediction Coding of Speech Signals

Language Modeling (I)

11/18

Language Modeling (I)

Acoustic Modeling (I):

11/25

Acoustic Modeling (II): Cambridge Hidden Markov Model Toolkit(HTK)

Homework 4:Exercises on HTK Toolkit (Due 12/2)

12/2

Acoustic Modeling (I): Triphone Modeling, CART etc.

Search Algorithms

Homework 5:Derive the equations of likelihood gains used for data splitting, on P. 179-180 of the textbook (Due 12/9)12/9

Invited Speaker: Roger Kuo (郭人瑋)

Acoustic Modeling (III): Adaptation Techniques for Acoustic Models

12/16

Invited Speaker: Louis Tasi (蔡文鴻)

Language Modeling (II): SRI Language Modeling Libraries and Tools

Language Modeling (III): Adaptation Techniques for Language Models

12/23

Search Algorithms

Large Vocabulary Continuous Speech Recognition (LVCSR)

12/30

Robustness Techniques for Feature Extraction

1/6

Final Exam

Discriminant Feature Extraction and Dimension Reduction

Spoken Dialogue Techniques

Textbook:

1. X. Huang, A. Acero, H. Hon, “Spoken Language Processing,” Prentice Hall, 2001 (全華代理)References:

Books:

1. T. F. Quatieri,“Discrete-Time Speech Signal Processing - Principles and Practice,” Prentice Hall, 2002

2. J. R. Deller, J. H. L. Hansen, J. G. Proakis, “Discrete-Time Processing of Speech Signals,” IEEE Press, 2000

3. F. Jelinek, "Statistical Methods for Speech Recognition," The MIT Press, 1999

4. S. Young et al., “The HTK Book”, Version 3.2, 2002. "http://htk.eng.cam.ac.uk"

5. L. Rabiner, B.H. Juang, “Fundamentals of Speech Recognition”, Prentice Hall, 1993

Papers:

1. Rabiner, “A Tutorial on Hidden Markov Models and Selected Applications in Speech

Recognition,” Proceedings of the IEEE, vol. 77, No. 2, February 1989

2. A. Dempster, N. Laird, and D. Rubin, "Maximum likelihood from incomplete data via the EM algorithm,"

J. Royal Star. Soc., Series B, vol. 39, pp. 1-38, 1977

3. Jeff A. Bilmes "A Gentle Tutorial of the EM Algorithm and its Application to Parameter

Estimation for Gaussian Mixture and Hidden Markov Models," U.C. Berkeley TR-97-021

4. J. W. Picone, “Signal modeling techniques in speech recognition,” proceedings of the

IEEE, September 1993, pp. 1215-1247

5. R. Rosenfeld, ”Two Decades of Statistical Language Modeling: Where Do We Go from

Here?,” Proceedings of IEEE, August, 2000

6. Hermann Ney, “Progress in Dynamic Programming Search for LVCSR,” Proceedings of the IEEE, August 2000

7. "Progress in Dynamic Programming Search for LVCSR", Proceedings of the IEEE, 88(8), August 2000.

8. H. Hermansky, "Should Recognizers Have Ears?", Speech Communication, 25(1-3), 1998.