Python module for form-sensitive transitional probability analysis of harmonic structures in corpora of songs/movements
A sample corpus is available from http://www.github.com/kshaffer/CCLI-2011
This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.
Most classes, methods, and functions are fairly self explanatory from looking at the source code, if you are familiar with Python and object-oriented programming. Even if you aren't, here are some sample code snippets for analyzing songs within the CCLI-2011 corpus that should make it pretty easy.
To import songs into a corpus:
from corpusAnalysis import *
CCLI = CorpusAnalysis()
CCLI.repertoire.append('01HowGreatIsOurGod.csv')
CCLI.repertoire.append('02MightyToSave.csv')
CCLI.repertoire.append('03OurGod.csv')
CCLI.repertoire.append('04BlessedBeYourName.csv')
CCLI.repertoire.append('05HereIAmToWorship.csv')
and so on...
To generate a transitional probability table for all songs in the corpus and write to CSV file:
results = CCLI.transitionalProbability()
f = open('./probTable.csv', 'ab')
w = csv.writer(f, delimiter=',')
for row in results:
w.writerow(row)
f.close()
The real power of this module lies in its ability to perform form-sensitive analyses. If a song is encoded with phrase boundaries (single pipes "|") and module boundaries (double pipes "||"), transition tallies and probability profiles can be calculated ignoring chords that cross either or both of those boundaries, using CORPUS.transitionalProbabilityNoPhraseBreaks() or CORPUS.transitionalProbabilityNoModuleBreaks().
CORPUS.transitionalProbabilityWithinModule(moduleName) and CORPUS.transitionalProbabilityWithinModuleNoPhraseBreaks(moduleName) can make calculations within song sections. For example, in a corpus of pop/rock songs, you can run one of these methods for verses, then choruses, then bridges and compare the results.
I trust that the rest of the methods and classes are pretty self explanatory once you understand the ones described here. Happy analyzing! And be sure to check out the CCLI-2011 corpus at github.com/kshaffer, both for something to play around with, and to see an example of the file format.
Though the example corpus uses Roman numerals, any nomenclature can be used, including functional bass, Riemannian functions, figured bass, etc. Just be consistent
Comments and questions welcome at kris.shaffer@gmail.com.