Digital Speech Processing Course (Fall 2009)

No Cheating Policy:

                No cheating declaration: (No Cheating Policy.pdf)

 Class Schedule: (Course_Outline_fall_2009.pdf)

Projects:

         General Range of Projects: (Digital Speech Processing Projects.pdf)

               LPC Vocoder Project: (LPC Vocoder Project.pdf)

Lectures:

Introductory Material:(basic course material_fall_2009.pdf)--6-to-a-page:(basic course material_fall_2009_6tp.pdf)

Lecture 1: (Introduction to Digital Speech Processing):(Lecture 1_fall_2009.pdf) -- 6 to a page:(Lecture 1_fall_2009_6tp.pdf)

        Book Chapter 1: (chapter1_6_3_2009.pdf)

Lecture 2:(Review of DSP Fundamentals):(Lecture 2_fall_2009.pdf)  -- 6 to a page: (Lecture 2_fall_2009_6tp.pdf)

        Book Chapter 2: (chapter2_6_3_2009.pdf)

Lecture 3:(Acoustic Theory of Speech Production):(Lecture 3_fall_2009.pdf)   -- 6 to a page: (Lecture 3_fall_2009_6tp.pdf)

        Pitch Detection Supplement: (Lecture 3_pitch_measurement.pdf)

        Book Chapter 3: (chapter3_6_17_2009.pdf)

Lecture 4:(Speech Perception--Auditory Models, Sound Perception Models, MOS Methods):(Lecture 4_fall_2009.pdf)  -- 6 to a page: (Lecture 4_fall_2009_6tp.pdf)

        Book Chapter 4:(chapter4_6_30_2009.pdf)

Lectures 5-6:(Sound Propagation in the Vocal Tract): (Lectures 5-6_fall_2009.pdf)   -- 6 to a page:  (Lectures 5-6_fall_2009_6tp.pdf)

        Book Chapter 5: (chapter5_6_23_2009.pdf)

Lectures 7-8:(Time Domain Methods in Speech Processing): (Lectures 7-8_fall_2009.pdf) -- 6 to a page: (Lectures 7-8_fall_2009_6tp.pdf)

        Book Chapter 6: (chapter6_6_3_2009.pdf)

Lecture 9:(Short Time Fourier Analysis Methods--Filter Bank Summation and Overlap Add): (Lecture 9_fall_2009.pdf)   -- 6 to a page: (Lecture 9_fall_2009_6tp.pdf)

Lecture 10:(Short Time Fourier Analysis Methods--Modifications and Filter Bank Design): (Lecture 10_fall_2009.pdf)  -- 6 to a page: (Lecture 10_fall_2009_6tp.pdf)

Lecture 11:(Speech Representations Based on STFT Analysis-Synthesis Methods):  (Lecture 11_fall_2009.pdf) -- 6 to a page: (Lecture 11_fall_2009_6tp.pdf)

        Book Chapter 7:  (chapter7_6_3_2009.pdf)

Lecture 12: (Homomorphic Speech Processing): (Lecture 12_fall_2009.pdf)  -- 6 to a page: (Lecture 12_fall_2009_6tp.pdf)

        Book Chapter 8:  (chapter8_6_3_2009.pdf)

Lecture 13: (Linear Predictive Coding Methods): (Lecture 13_fall_2009.pdf)  -- 6 to a page: (Lecture 13_fall_2009_6tp.pdf)

        Book Chapter 9:  (chapter9_6_3_2009.pdf)

Lecture 14:(LPC Methods--Lattice Structures):  (Lecture 14_fall_2009.pdf) -- 6 to a page: (Lecture 14_fall_2009_6tp.pdf)

Lecture Algorithms: (Speech Detection, VUS Decision, Pitch Detection, Formant Estimation): (Lecture_algorithms_fall_2009.pdf)   --6 to a page:(Lecture_algorithms_fall_2009_6tp.pdf)

        Book Chapter 10:(chapter10_6_23_2009.pdf)

Lecture 15:(Speech Coding Based on Speech Waveform Representations and Speech Models): (Lecture 15_fall_2009.pdf) -- 6 to a page: (Lecture 15_fall_2009_6tp.pdf)

Lecture 16: (Speech Coding--Adaptive and Differential Methods):(Lecture 16_fall_2009.pdf)  -- 6 to a page: (Lecture 16_fall_2009_6tp.pdf)

Lecture 17:(Speech Coding--Model-Based Coding and Speech Coding Standards): (Lecture 17_fall_2009.pdf)  -- 6 to a page: (Lecture 17_fall_2009_6tp.pdf)

        Book Chapter 11: (chapter11_6_3_2009.pdf)

Lecture 18: (Text-to-Speech Synthesis (TTS) Methods): (Lecture 18.pdf)  -- 6 to a page: (Lecture 18_6tp.pdf)

        Book Chapter 12: (chapter13_6_3_2009.pdf)

Lecture 19:(Techniques for Speech Recognition and Natural Language Understanding): (Lecture 19.pdf)  -- 6 to a page: (Lecture 19_6tp.pdf)

        Book Chapter 13: (chapter14_6_3_2009.pdf)

Lecture 20:(The Hidden Markov Model--HMM): (Lecture 20.pdf)  -- 6 to a page: (Lecture 20_6tp.pdf)

 


Homeworks:

 

Problem Set 1: (ps1_fall09.pdf),    PS1 solution: (ps1_soln_fall09.pdf)

Problem Set 2: (ps2_fall09.pdf),    PS2 solution:  (ps2_soln_fall09.pdf)

Problem Set 3: (ps3_fall09.pdf),    PS3 solution: (ps3_soln_fall09.pdf)

Problem Set 4: (ps4_fall09.pdf) ,   PS4 solution:  (ps4_soln_fall09.pdf)

Problem Set 5: (ps5_fall09.pdf) ,   PS5 solution: (ps5_soln_fall09.pdf)

Problem Set 6: (ps6_fall09.pdf) ,   PS6 solution: (ps6_soln_fall09.pdf)

Problem Set 7: (ps7_fall09.pdf) ,   PS7 solution: (ps7_soln_fall09.pdf)

Problem Set 8: (ps8_fall09.pdf) ,   PS8 solution: 

 


Speech Files:

 

test_16k.wav: (test_16.wav)

ah.wav: (ah.wav)

should.wav: (should.wav)

s3.wav: (s3.wav)

s5.wav: (s5.wav)

s5_synthetic.wav: (s5_synthetic.wav)

we_were: (we were away a year ago_lrr.wav)

s3_pitch_file: (pp3.mat)

s5_pitch_file: (pp5.mat)

s1.wav: (s1.wav)

s1_pitch_file: (pp1.mat)

s2.wav: (s2.wav)

s2_pitch_file: (pp2.mat)

s4.wav: (s4.wav)

s4_pitch_file: (pp4.mat)

s6.wav: (s6.wav)

s6_pitch_file: (pp6.mat)

 

beep1:(beep_fs_8000.wav)

beep2:(beep_fs_10000.wav)

beep3:(beep_fs_16000.wav)

 

tidigits files:

1:(1A.waV), (1B.waV); 2:(2A.waV), (2B.waV); 3:(3A.waV), (3B.waV); 4:(4A.waV), (4B.waV); 5:(5A.waV), (5B.waV)

6:(6A.waV), (6B.waV); 7:(7A.waV), (7B.waV); 8:(8A.waV), (8B.waV); 9:(9A.waV), (9B.waV); oh:(OA.waV), (OB.waV)

zero:(ZA.waV), (ZB.waV)

tidigits training set, endpointed: (isolated_digits_ti_train_endpt.zip)

tidigits testing set, endpointed: (isolated_digits_ti_test_endpt.zip)

 

cepstral coefficient files:

1:(cc_tidig_endpt_1.mat), 2:(cc_tidig_endpt_2.mat), 3:(cc_tidig_endpt_3.mat), 4:(cc_tidig_endpt_4.mat)

5:(cc_tidig_endpt_5.mat), 6:(cc_tidig_endpt_6.mat), 7:(cc_tidig_endpt_7.mat), 8:(cc_tidig_endpt_8.mat)

9:(cc_tidig_endpt_9.mat), oh:(cc_tidig_endpt_O.mat), zero:(cc_tidig_endpt_Z.mat)

 

dtw train and test files:

(train.mat), (test.mat)

 

template files for 11 digits (1-9,oh,zero):

1:(template_isodig_1.mat), 2:(template_isodig_2.mat), 3:(template_isodig_3.mat), 4:(template_isodig_4.mat)

5:(template_isodig_5.mat), 6:(template_isodig_6.mat), 7:(template_isodig_7.mat), 8:(template_isodig_8.mat)

9:(template_isodig_9.mat), oh:(template_isodig_10.mat), zero:(template_isodig_11.mat)

 

lrr digit training  and testing files (endpointed files):

    training set: (train.zip);  testing set: (test.zip)

    list of training files: (files_lrrdig_isodig_train_endpt.mat)

    list of testing files: (files_lrrdig_isodig_test_endpt.mat)

 

lrr digit training and testing files (unendpointed files):

    training set:(digits_lrr_train_orig.zip)

    testing set: (digits_lrr_test_orig.zip)

 


Matlab Files:

 

loadwav.m: (m file)

savewav.m: (m file)

loadraw.m: (m file)

saveraw.m: (m file)

 

strips_modified.m:(strips_modified.m)

QMF Designs: (filters)

 

grayscale.m: (m file)

fxquant.m: (m file)

pspect.m: (m file)

play_file.m: (m file)

 

HW7 LPC Analysis code: (test_lpc.m)

autocorrelation method code: (autolpc.m)

durbin solution code: (durbin.m)

cholesky solution code: (cholesky_full.m), (cholesky.m)

lattice solution code: (lattice.m)