Borne out of a growing frustration with the lack of progress on the
semester projects, the instructor assigned two one hour programming
tasks shown below.  This turned into a long ordeal - only 25% of the
students finished the assignments within one week...

Several students' solutions (correct solutions, of course) to these
problems are provided in these directories.

-Joe Picone

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Project No. 1: Compute the SNR of a speech file

Date: Tue, 13 Feb 96 22:38:35 -0600
From: Joe Picone - The Terminal Man <picone@isip01.isip.msstate.edu>
To: ee_8993
Subject: quiz #1
Reply-to: picone@isip.msstate.edu

I am tired of seeing so many vacuous faces in class when I throw
up equations. Hence, I am forced to resort to quizzes :)

By Feb. 22, I would like each of you to produce a program to compute
the SNR of a speech file. The program should run from the command line
as follows:

	cat speech_file | my_prog.exe

The output should be:

	the SNR is XX.XX dB

The program should do the following:

(1) assume 8 kHz data
(2) preemphasize using 1 - 0.95 z^-1
(3) use a frame duration of 20.0 msec / window duration of 30.0 msec
(4) hamming window the data
(5) nominal signal level is the 85%; the nominal noise level is 15%

You MUST work alone - no collaborations. If your code even looks
similar to someone else's code, your new assignment will be to
build a real-time recognizer for ARPA's NAB task (about 100K words)!

This is worth some non-negligible percentage of your grade to be determined
during your presentation of your project (the angrier I get...)

- The "I don't think my students know how to program mathematics" Terminal Man

-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Project No. 2: Compute the spectrum of an LPC model

Date: Sat, 2 Mar 96 08:22:49 -0600
From: Joe Picone - The Terminal Man <picone@isip01.isip.msstate.edu>
To: ee_8993
Subject: Quiz #2
Reply-to: picone@isip.msstate.edu

As discussed in class, here is the second short computer problem I
would like you to solve:

Create a program that functions as follows:

 cat foo_8kHz.raw | my_prog.exe preemphasis window_duration center_time lp_order | xmgr -source stdin

The net result should be a plot of the signal spectrum computed
using the following parameters:

 fs = 8 kHz
 preemphasis = argv[1]
 window_duration = argv[2]
 center time of the window = argv[3]
 hamming window = yes
 lp_order = argv[4]

and plotted on a log amplitude vs. linear frequency scale. The spectra
of the corresponding linear prediction model should be plotted as
well. Xmgr accepts multiple sets of data, so simply print your xy points
for both plots to stdout, with the second set separated by a newline,
and xmgr will take care of the rest.

You can use a DFT to compute the spectrum of the signal, or a zero-stuffed
fft. The important thing is to only use window_duration number of samples
of real data (note that window_duration is specificed in secs).

For example,

 cat foo_8kHz.raw | my_prog.exe 0.95 0.03 3.0 12 | xmgr -source stdin

should produce a signal and lp model spectrum for a 30 msec window of the
signal centered at 3 secs. The lp analysis will be of order 12. A preemphasis
filter 1 - 0.95z**-1 is applied to the data.

For most of you, this should be a useful tool to have around. Feel
free to pull the LP analysis software off of the net. The main thing
is to get the visualization component working - and to understand
gain matching of the two spectra.

The resulting plots will typically have about a 60 dB dynamic range
for studio quality data.

This assignment is due March 11 or sooner.

Also, please remember to send me your "finished" abstracts.

-Joe