Difference between revisions of "Music 253 Humdrum homework"
(13 intermediate revisions by the same user not shown) | |||
Line 1: | Line 1: | ||
+ | |||
+ | __NOTOC__ | ||
Do some vocal-range analysis of J.S. Bach chorales. | Do some vocal-range analysis of J.S. Bach chorales. | ||
Line 4: | Line 6: | ||
See [[Getting started with Humdrum]] to access the Humdrum Tools. | See [[Getting started with Humdrum]] to access the Humdrum Tools. | ||
− | You can download the Bach chorales by doing these commands in a terminal | + | You can download the Bach chorales by doing these commands in a terminal: |
mkdir chorales | mkdir chorales | ||
cd chorales | cd chorales | ||
Line 13: | Line 15: | ||
or with more detail such as the size of the files: | or with more detail such as the size of the files: | ||
ls -asF | ls -asF | ||
+ | |||
+ | If you are using Virtual Humdrum, then the chorales are already downloaded: | ||
+ | cd data/bach-js/371chorales/kern | ||
+ | ls -asCF | ||
Line 27: | Line 33: | ||
− | + | Intermediate output results can either be saved to a file, and then processed with prange, or the data can be sent directly to prange: | |
extractx -f 1 chor001.krn > temp-file | extractx -f 1 chor001.krn > temp-file | ||
prange temp-file | prange temp-file | ||
Line 41: | Line 47: | ||
Load the choral.pmx file into SCORE with the command: | Load the choral.pmx file into SCORE with the command: | ||
RE chor001.pmx | RE chor001.pmx | ||
+ | |||
+ | |||
+ | Alternatively, you can convert the PMX data into an SVG image from the command line with this program: | ||
+ | |||
+ | #!/usr/bin/perl | ||
+ | |||
+ | use HTTP::Request::Common; | ||
+ | use LWP::UserAgent; | ||
+ | |||
+ | my $data; | ||
+ | my $line; | ||
+ | while ($line = <>) { | ||
+ | $data .= "$line"; | ||
+ | } | ||
+ | |||
+ | $ua = LWP::UserAgent->new; | ||
+ | my $response = $ua->request( | ||
+ | POST 'http://score.sapp.org/cgi-bin/score', | ||
+ | [ | ||
+ | outputformat => 'svg', | ||
+ | embedpmx => 'yes', | ||
+ | inputdata => [$data], | ||
+ | ] | ||
+ | ); | ||
+ | |||
+ | if ($response->is_success) { | ||
+ | print $response->decoded_content; | ||
+ | } else { | ||
+ | die $response->status_line; | ||
+ | } | ||
+ | |||
+ | If this PERL script is saved to a file called "getsvg", then to generate the SVG image: | ||
+ | |||
+ | ./getsvg file.pmx > file.svg | ||
+ | |||
+ | [note: you should run the following commands to make the getsvg script executable]: | ||
+ | chmod 0755 getsvg | ||
+ | |||
+ | All in one step: | ||
+ | humcat chor001.krn | prange --score | ./getsvg > chor001.svg | ||
== Question 3 == | == Question 3 == | ||
Line 63: | Line 109: | ||
== Question 5 == | == Question 5 == | ||
− | What is the average twelve-tone pitch | + | What is the average twelve-tone pitch for each voice? What is the lowest quartile and highest quartile pitch for each voice? The lowest quartile point is where 25% of notes are below and 75% of notes are above. 50% of notes fall between the 25% and 75% quartile points. You could also generate a box-and-whisker plot by hand or with a program: |
+ | |||
+ | |||
+ | http://www.shmoop.com/basic-statistics-probability/box-whisker-plots.html | ||
+ | |||
+ | https://www.youtube.com/watch?v=8vul6_7TlWY | ||
+ | https://www.youtube.com/watch?v=Cj-buDcHVMw | ||
== Question 6 == | == Question 6 == | ||
− | + | What is the 2-standard deviation of each voice's pitch range (the range of notes which covers 95% of the range that they sing). | |
This should be the Tessitura (https://en.wikipedia.org/wiki/Tessitura) of the vocal ranges. How does the measured tessitura match to standard listings for vocal ranges of soprano, alto, tenor, bass? | This should be the Tessitura (https://en.wikipedia.org/wiki/Tessitura) of the vocal ranges. How does the measured tessitura match to standard listings for vocal ranges of soprano, alto, tenor, bass? |
Latest revision as of 08:11, 12 March 2016
Do some vocal-range analysis of J.S. Bach chorales.
See Getting started with Humdrum to access the Humdrum Tools.
You can download the Bach chorales by doing these commands in a terminal:
mkdir chorales cd chorales humsplit h://chorales
To see a list of the chorales:
ls
or with more detail such as the size of the files:
ls -asF
If you are using Virtual Humdrum, then the chorales are already downloaded:
cd data/bach-js/371chorales/kern ls -asCF
Question 1
Use the Humdrum Extras programs 'extractx' and 'prange' to calculate the number of notes each voice plays on each twelve-tone pitch in chor001.krn. Plot the pitch range data as a histogram for each voice in a plotting program of your choice.
Here is how to extract the first part from a file:
extractx -f 1 chor001.krn
Or to explicitly extract by instrument name:
extractx -g Bass chor001.krn
Intermediate output results can either be saved to a file, and then processed with prange, or the data can be sent directly to prange:
extractx -f 1 chor001.krn > temp-file prange temp-file
extractx -f 1 chor001.krn | prange
Question 2
Try the “– –score” option for prange to generate SCORE data for pitch range histograms for each voice in chor001.krn. Print the SCORE output from prange for chor001.krn and a composite of all chorales.
humcat chor001.krn | prange --score > chor001.pmx
Load the choral.pmx file into SCORE with the command:
RE chor001.pmx
Alternatively, you can convert the PMX data into an SVG image from the command line with this program:
#!/usr/bin/perl use HTTP::Request::Common; use LWP::UserAgent; my $data; my $line; while ($line = <>) { $data .= "$line"; } $ua = LWP::UserAgent->new; my $response = $ua->request( POST 'http://score.sapp.org/cgi-bin/score', [ outputformat => 'svg', embedpmx => 'yes', inputdata => [$data], ] ); if ($response->is_success) { print $response->decoded_content; } else { die $response->status_line; }
If this PERL script is saved to a file called "getsvg", then to generate the SVG image:
./getsvg file.pmx > file.svg
[note: you should run the following commands to make the getsvg script executable]:
chmod 0755 getsvg
All in one step:
humcat chor001.krn | prange --score | ./getsvg > chor001.svg
Question 3
Calculate and plot pitch histograms for each voice (soprano, alto, tenor, bass) in all 370 chorales. One histogram for each voice (not one for each chorale).
Extract all notes from all chorales like this (for Bass part):
extractx -f 1 chor*.krn | prange
Also generate and print with SCORE similar plots:
prange chor*.krn --score
Optional: add mouse hovering code to list the number of notes for each pitch
prange chor*.krn --score --hover
Question 4
Based on the data/plots from question 3, what is the maximum and minimum pitch for each part in all of the chorales?
Question 5
What is the average twelve-tone pitch for each voice? What is the lowest quartile and highest quartile pitch for each voice? The lowest quartile point is where 25% of notes are below and 75% of notes are above. 50% of notes fall between the 25% and 75% quartile points. You could also generate a box-and-whisker plot by hand or with a program:
http://www.shmoop.com/basic-statistics-probability/box-whisker-plots.html
https://www.youtube.com/watch?v=8vul6_7TlWY
https://www.youtube.com/watch?v=Cj-buDcHVMw
Question 6
What is the 2-standard deviation of each voice's pitch range (the range of notes which covers 95% of the range that they sing).
This should be the Tessitura (https://en.wikipedia.org/wiki/Tessitura) of the vocal ranges. How does the measured tessitura match to standard listings for vocal ranges of soprano, alto, tenor, bass?
https://en.wikipedia.org/wiki/Soprano
https://en.wikipedia.org/wiki/Alto