Tabulate the consonant to vowel phoneme ratios of languages in the UCLA Phonological Segment Inventory Database (UPSID).

The data files for this project can be found in the UPSID-PC distribution, which is available for download here:

http://phonetics.linguistics.ucla.edu/sales/software.htm

Within the ZIP file for the program, the data files are:

• CHARCOD.STR
• FEATS.STR
• LANGS.STR
• SEGS.STR

The subfolders UPSID2 UPSID3 and UPSID4 contain reference files, with one file per language. The file names in those directories include the numeric language ID code that is used within UPSID. The files are plain-text files that identify the language and reference the sources that were used to derive the phonological data for the language that appears within UPSID.

A simple web interface to the UPSID data is available here:

http://web.phonetik.uni-frankfurt.de/upsid.html

This web interface is not required for this CVRatio project, but a link is provided here for convenience to the reader.

To generate the filtered list of languages, you will also need a table of ISO-639-1 language codes. You can get the table from the following website:

https://www.loc.gov/standards/iso639-2/

This is the website for ISO-639-2, but ISO-639-1 is a subset of ISO-639-2 and so its data is included within. Follow the ISO 639-2 Code List link, and then the Code list for downloading link. You want the Character encoding in UTF-8 version.

The result should be a plain-text file with records that use | as a field separator, where each record has ISO 639-2 three-letter codes, an ISO 639-1 two-letter code if available, and the language names in both English and French.

The complete tabulation of UPSID languages according to number of consonant phonemes, number of vowel phonemes, and ratio of consonants to vowels is given in results-full.txt. The records are sorted in descending order of the ratio of consonants to vowels.

results-subset.txt is a subset of results-full.txt that only includes languages that have a two-letter language code in ISO-639-1. This filters many rare languages out of the UPSID data, since data for rare languages is often unreliable.

The scripts used to generate the results are in the src subdirectory of this project. See the README.md file in that subdirectory for further information.

The intermediate data files used to generate the results are in the data subdirectory of this project. See the README.md file in that subdirectory for instructions how to generate the intermediate data files and how to generate the results-full.txt and results-subset.txt data files.