sindhi

Codebase for Indic-Transliteration using Seq2Seq RNN. For latest repo with Transformer-based models, check: https://github.com/AI4Bharat/IndicXlit

RTL fixes for Arabic-script languages in Codex through ChatGPT Desktop on Windows

PALI: Language identification for Perso-Arabic Scripts

Indo-Pakistani Transliteration System [WIP 🚧]

A Flutter library to add multi lingual keyboards in Flutter App

Script Normalization for Unconventional Writing of Perso-Arabic scripts (ACL2023)

🧠️🖥️2️⃣️0️⃣️0️⃣️1️⃣️🔠️🔢️ The linguistic:Sindhi category for AI2001, containing Sindhi language linguistic datasets

Python project for scraping Daily Sobh news data and training machine learning (Logistic Regression, Decision Tree, Random Forest, SVM) and LSTM models to predict news topics from headlines. Developed for the DSA course at SMIU. Disclaimer: Educational purposes only. Not affiliated with Daily Sobh. Contributor: Furqan Hun.

Sindlish — a high-level programming language for the Sindhi-speaking community, with a bytecode VM backend.

A Small Language Model for the Sindhi Language (Work in Progress)

Nastaliq Fonts for Slackware Linux

Large-Scale Font-Diverse Sindhi Ligature Recognition Datasets and Models

An offline NLP toolkit for Sindhi (Perso-Arabic script) featuring Unicode normalization, tokenization, stopword removal, rule-based stemming, lexicon-based sentiment analysis, and text statistics with an interactive Streamlit interface.

A modern digital implementation of the ancient Sindhi board game Bhariyo.

Open-source patcher that adds RTL text support and the Vazirmatn font to Claude Desktop, for Persian, Arabic, Hebrew, Urdu, and other RTL languages

Extra Devanagari fonts for Slackware Linux

Sentiment classifier for Sindhi text (dual TF-IDF + LinearSVC) — v2: 91.7% acc / 0.918 macro F1 on a fully human-labeled 4,420-sentence dataset. Includes a transparent writeup of why accuracy dropped from v1's biased eval set.

Extra Arabic fonts for Slackware Linux

Village-level drinking-water accessibility for 5,159 villages in Sindh, Pakistan, from open satellite data, and why the standard method ranks the driest district as best-served.