Onset

Onset is a language evolution simulator, which evolves a list of words in IPA form according to realistic phonological rules.

The frontend is built with Vue and the CSS framework Bulma. It communicates with the backend using simple REST endpoints.

The backend is built with Python and Flask.

Directory Structure

• src is the source code for the Vue frontend
• app is the source code for the Python API, written in Flask.
• engine is the source code for the evolution engine which is called by Flask.
• app/templates/ contains the Webpack-generated index file, served with Flask.
• app/static/ contains static assets generated by Webpack.
• config contains Webpack configuration files, generated using vue-cli

Build Setup

# install Python dependencies
pip install -r requirements.txt

# install Javascript dependencies
npm install

# build frontend
npm run build

# run using Flask's development server
python run.py

# or use PyPy for a speed boost
pypy3 run.py

To install development requirements, which will allow testing, validation, and script usage:

pip install -r requirements-dev.txt

To validate the YAML data:

pykwalify -d engine/data/rules.yaml -s engine/data/rules.schema.yaml
pykwalify -d engine/data/diacritics.yaml -s engine/data/diacritics.schema.yaml

To run the tests:

py.test

Sources

A variety of sources were used for the information needed to build this app. Please see the LICENCE.md file in the engine/data directory for specific data sources.

The following papers were used when implementing the algorithms:

• Harold R. Bauer (1988) The ethologic model of phonetic development: I. Phonetic contrast estimators, Clinical Linguistics & Phonetics, 2:4, 347-380, DOI: 10.3109/02699208808985265
• Carol Stoel-Gammon (2010) The Word Complexity Measure: Description and application to developmental phonology and disorders, Clinical Linguistics & Phonetics, 24:4-5, 271-282, DOI: 10.3109/02699200903581059
• Carterette, E. and Jones, M. (1974) Informal Speech: Alphabetic and Phonetic Texts with Statistical Analyses and Tables (Berkeley: University of California Press).

General information on the linguistics underpinning the app is from:

• Trask's Historical Linguistics by Larry Trask.
• Introductory Phonology by Bruce Hayes.
• Wikipedia

A lot of technical inspiration was taken from the source code of panphon. In particular, the deparsing algorithm and YAML data files were inspired by panphon's approach. Please check it out!