2023 – present · Ongoing project
NLP Kurdî is a project of the Mesopotamia Foundation (Weqfa Mezopotamyayê). We started the project in 2023, and the work is still ongoing. Our aim is to give Kurdish Kurmanji the everyday language technology that widely used languages already have, so that people can write, read, listen to and speak Kurmanji with digital tools.
The platform brings together two areas of work:
Text and grammar tools
These tools run on a morphological system that describes how Kurmanji words are formed and inflected. This part draws heavily on resources our Foundation had already prepared: our books, the spelling dictionary Ferhenga Rastnivîsînê and the spelling guide Rêbera Rastnivîsînê.
- Spell Checker (Rastnûser) — finds and corrects spelling mistakes in Kurmanji text
- Verb Conjugator (Karkêş) — conjugates Kurmanji verbs in every tense, person and mood
- Word Analyzer — breaks a word down into its stem and morphemes
- Text Analyzer — analyzes every word in a longer text
- Noun/Adjective — shows the inflected forms of nouns and adjectives
- Rêbera Rastnivîsînê — the Kurmanji spelling guide
Speech technology
These tools are built on Kurmanji speech and text data collected and prepared for the project, and on models trained on that data.
- Speech to Text (Dengnûs) — turns spoken Kurmanji into written text
- Text to Speech (Deqxwîn) — reads Kurmanji text aloud
All of this was made possible by the dedicated team listed below. They built the linguistic system, prepared the datasets, trained the models, developed the backend services that run them, and built the website and server infrastructure that make the tools freely available online.
Project Team
- Mehdî Jafarzadeh (Project Coordinator)
- Aso Mehmudî
- Osman Akyil
Linguistic Consultant
Dr. Mîkaîl Bulbul
The Mesopotamia Foundation sincerely thanks the project team for their dedication and hard work, and Dr. Mîkaîl Bulbul for their valuable advice. Their contribution helps keep Kurmanji alive in the digital world.
