Department of Electrical and Computer Engineering IIT Kanpur

EE627A - Jul. 2018

Speech Signal Processing

Home page

Content

Announcements

Schedule/ Lecture Slides

Student Presentations

Grading Policy

Class Notes - Contact the TAs for copies of class notes.

Lecture 1/2

a) Notes on Sampling and Reconstruction (scanned PDF)
b) pp.72 - pp.78 of Thomas Quateiri Discrete Time Speech Signal Processing Prentice Hall
c) pp.11-36 of Chapter 2 : Rabiner and Juangs Book
d) Lecture Slides (PDF)
e) pp.79-pp.94 of Thomas Qauteiri's book.

Lecture 3/4

a) Lecture Slides (PDF)
b) Additional Material as PDF
c) Chapter 1 (complete), and Chapter 2 (complete), T Quateiri, Discrete Time Speech Signal Processing
d) Chapter 3, pp.98-pp.106 Only, Digital Processing of Speech Signals, Rabiner and Schafer

Lecture 5/6
a) Lecture Slides (PDF)
b) Chap. 4 of Rabiner and Schafer pp.116-150;
c) Chap.6 of Rabiner and Schafer pp.250-266
d) Chap. 3 of Rabiner, Juang, and Yegna (Indian Edition) pp. 60-86

Lecture 7/8
a) Lecture Slides (PDF)
b) Chapter 4, Section 4.1 (pages 149 - 150 only), Section 4.2 (pages 156 - 160 only), Manolakis, Ingle, Kogon - Statistical and Adaptive Signal Processing (Blue Hard Cover Book)
c) Chapter 3 - pages 88 - 123 only), Rabiner, Juang and Yegna, Fundamentals of Speech Recognition
d) Optional reading - Chapter 8 of Rabiner and Schafer, pp.396-412

Lecture 11/12
a) Lecture Slides (PDF)
b) Reference matlab code (script file)
c) Chapter 6 : Deller and Hansen's book (pp. 352 -- 374, and pp. 385--401)

Lecture 13/14
a) Lecture Slides (PDF)
b) Picone's paper on Signal modeling (pp.1215 - 1233 only)
c) Chapter 6 (pp. 374 - pp.385 only), Deller, Proakis, Hansen, Discrete time processing of speech signals, Wiley
d) pp. 295 - pp.303 only of Morgan and Gold (optional)

--------------------------------------------------------------------------------------------------------------------------------------
Related Matlab Resources
e) Reference Matlab code

f) Matlab Toolboxes required for solving the Assignment 4 questions (Click on the links below to download the zipped versions of the toolboxes). Make sure you acknowledge the authors of these toolboxes if you would like use them later in you research.
g) Voicebox : MFCC, PLP ...
h) M Slaney's auditory toolbox : MFCC ...
i) D Ellis's rastamat : for RASTA PLP .... : This ones a tgz, so download first to your comp and then unzip.
j) Download the wav file sa1.wav by clicking on this LINK.
k) Also take a look at Dan Ellis's Matlab resources for feature extraction on the web @ this LINK and also this LINK
-----------------------------------------------------------------------------------------------------------------------------------------
Lecture 15/16

a) Lecture Slides (PDF)
b) Chapter 4 (pp. 141 - pp.170, pp. 122 - pp.130, pp. 184 - pp. 189 only), Rabiner and Juang, Fundamentals of Speech Recognition

Lecture 17/18

i) Lecture Slides (PDF)
ii) Additional Notes
iii) Chapter 14 (pp. 709 - pp. 724), T Quateiri, Discrete Time Speech Signal Processing
Additional references (not mandatory for quiz and finals) if you are interested in knowing more
iv) Reynolds paper on speaker identification

Lecture DTW

i) Slides will be mailed to all registered students as and when the topic is discussed.
ii) Chapter 11, pp.623--650, of Deller and Hansen's book

Lecture HMM 1

Hidden Markov Models for ASR - Part I
i) Lecture Slides
ii) Chapter 6 (pp. 321 - pp. 352), Rabiner and Juang, Fundamentals of Speech Recognition

Lecture HMM 2

Hidden Markov Models for ASR - Part II
i) Lecture Slides
ii) Chapter 6 (pp. 321 - pp. 352), Rabiner and Juang, Fundamentals of Speech Recognition

Rajesh Hegde<rhegde@ucsd.edu>
Dept. of Electrical Engg. IIT Kanpur