orthography

The open, machine-readable standard for the Somali language — orthography, grammar, terminology, translation, and AI/benchmark resources, as a versioned, citable RFC-style catalog.

这是一个潮州话文本端的处理工具和正字标准，主要为潮州方言的语音合成服务

lo vrici se pilno be fi la zbalermorna je lo ri taipfase

Better Estonian from your AI agent. Offline MCP server (no third-party calls) wrapping EstNLTK, EKI Reeglid + Riigi Teataja. 27 read-only tools: spelling, morphology, orthography, kantseliit, legal. Lean results. Claude, ChatGPT, Codex: one click or a URL. 99.1% on inflection_et, 100% once its 13 errors are fixed; 99.3% on 11k real rows.

Text and orthography toolkit for 869 African languages: grapheme-to-phoneme (native spelling or IPA), conversion between any two languages' orthographies, one shared cross-language spelling, and a reversible plain-ASCII form. Non-Latin scripts romanised. Pure Python, no dependencies.

Belarusian orthography converter - NPM package

Belarusian orthography converter - Web App

Set of helper string manipulation functions for handling Serbian language orthography rules correctly

Полный конспект по русскому языку для ЦТ/ЦЭ 2027. Правила, ловушки, исключения. В «Режиме редактирования» вы можете менять коспект под себя. Все правки сохраняются только на вашем устройстве, а оригинальный сайт останется неизменным.

Rigorously defined orthographies for text in polytonic ancient Greek.

Python package containing functions that may be useful for word research.

Convert Kalaallisut/Greenlandic to IPA (International Phonetic Alphabet); Convert old Greenlandic orthography (Kleinschmidt) to new (1973)

This package is a wrapper for the LanguageTool API (https://languagetool.org/http-api/#/default)

textlint rule that flags inappropriate transliterations of Ukrainian place and person names and auto-fixes them to the correct forms. Promotes respectful, consistent naming across documentation.

A handy tool to quickly look-up words in Woordenlijst.org within your active Windows-app

Open-source Automatic Speech Recognition (ASR) pipeline for Bashkir (Bashkort), Kazakh, and Kyrgyz languages with deterministic orthography correction.

English, but worse.

English and Dutch rhyming datasets (5k word pairs each) for five types of rhymes. Three open-source LLMs (Llama2, Llama3, CrystalChat) are tested on these datasets, with prompt variation.

Determine homoglyphs for utf-32 codepoints