rita — Python port of RiTa

RiTa is a library for natural language and generative writing: syllables, phonemes and stresses, parts of speech, conjugation, pluralization, stemming, letter-to-sound mapping, rhyming and alliteration, plus n-gram generation and the RiScript scripting language.

This is the Python port of RiTa, which is written in JavaScript and Java. The API mirrors the JavaScript one, so JS docs apply.

Install

pip install rita-py

Usage

from rita import RiTa

RiTa.tokenize("I'm sure he'll be interested")
# ["I'm", "sure", "he'll", "be", "interested"]

RiTa.syllables("elephant")       # "eh/l-ah/f-ah-n-t"
RiTa.phones("elephant")          # "eh-l-ah-f-ah-n-t"
RiTa.rhymes("cat")               # ["bat", "begat", "brat", "chat", ...]
RiTa.pos("The dog ran")          # ["dt", "nn", "vbd"]

# Generation
markov = RiTa.RiMarkov(2)
markov.add_text("the big dog and the big cat and the small dog")
markov.generate1()               # "the big cat and the big cat and the small dog"

# RiScript
RiTa.evaluate("the [big | small] [dog | cat]")   # "the small dog"

Notes

• Pure Python, no runtime dependencies. Python 3.10+.
• The canonical dictionary (rita_dict.json) is bundled with the package.
• Licensed under GPL-3.0, same as the JavaScript and Java implementations.

Development

The build is managed with uv:

uv sync              # create .venv and install dev dependencies
uv run pytest -q     # run the test suite

The test suite under tests/ is auto-generated from the JavaScript tests in ../js/test/ by ../scripts/js-to-py-tests.js — regenerate rather than editing test_*.py by hand. Some tests read fixture files from ../js/test/, so keep that checkout present when running the suite.