Chapter 6, figure 2 has "1000" along the x-axis twice. The second occurence should have been "1500".
Chapter 6 has example DFT calculations that show a matrix. With Xm values on the left, and xn values on the right, the matrix shows terms like e^{j2pi49/8}. For the DFT, there should be a minus sign in front of the j terms. This matrix does represent the IDFT, though the Xm and xn vectors should be switched.
Chapter 10 has this example code:
% create an example sound
t = 0:0.0001:3;
x = 0.9*cos(2*pi*440*t);
This is written to a .wav file, with fs specified as 8000.
But 1/8000 = 0.000125, so the sampling rate should have been 10,000.
Instead, the note written to the .wav file is 352 Hz, not the expected 440 Hz.
If we use 0.000125 as the sampling period, the fs would be correct at
8000 samples/second.
Appendix E, answer to Question 5.11 (page 461) has (42+6)/3 reduce down to 16.67 kHz. Of course this should be 16 kHz. It should read "so fs in the range of 16 to 18 kHz is OK."
Also, m is specified as 2, when the way the question is stated it should be 1. The question should be re-worded to say replications "between 0 and +18 kHz". This applies to two other questions in chapter 5: 5.13 and 5.22.
For consistency with the above note, question 5.13 should say "(i.e., between 0 Hz and +10 MHz)". Also, question 5.22 should read "one replica of the signal (between 0 Hz and +24 kHz)".
Appendix E, answer to Question 5.15 part c (page 462)
has numbers -20, -10, +20, and +10
where it should have -25, -15, +25, and +15.
It should read as follows:
"The 5 kHz frequency has replications every +/- 10 kHz,
i.e., at -25 kHz, -15 kHz,
-5 kHz, 15 kHz, 25 kHz, etc. Each of these
also appears mirrored around 0 Hz, i.e., +25 kHz,
+15 kHz,
+5 kHz, -15 kHz, -25 kHz, etc."
Appendix E, page 464, question 21: "100 - 44 - 44 = 22" should be "100 - 44 - 44 = 12".
Click here for errata on slides.
Having trouble with the compress_test program? Look here for a solution.