Awesome Kurdish

(last updated on 04/01/2023)

A curated list of awesome resources, tools and scientific papers for Kurdish language technology

Although I do my best to keep this page as comprehensive as possible by including all projects, the list may not include all the fantastic small and big projects regarding Kurdish language processing. Please be kind and notify me by reaching out by email or through our community on Gitter.

Are you interested in contributing to Kurdish language processing? Check out this post to see how you can do so.

Development

Resources

Corpora

• Open Super-large Crawled ALMAnaCH coRpus (OSCAR) (Sorani and Kurmanji)
• Pewan (Sorani and Kurmanji)
• Kurdish folkloric lyrics corpus (Sorani)
• AsoSoft corpus (Sorani)
• Kurdish Textbooks Corpus (Sorani)
• Zaza-Gorani corpus (Zazaki and Gorani)
• Kurdish resources on Clarin
• University of Bamberg's corpora [Kurmanji & Laki]

Parallel corpora

• Ataman's Bianet corpus containing Turkish-English-Kurmanji aligned texts
• Ahmadi et al's corpus containing English-Kurmanji-Sorani aligned texts
• Tanzil: one Qoran translation alignable with many other translations in other languages, including 11 in English (see this project)
• Bible translations in Kurmanji-Latin and Kurmanji-Cyrillic
• TED Talks subtitles
• HLP Colloquial Corpus #1 (Sorani and Kurmanji (Latin and Arabic)) (not free)
• A parallel corpus of Sorani-English text
• FLORES-101 Evaluation Benchmark for Low-Resource and Multilingual Machine Translation (Sorani)
• AsoSoft Speech Corpus for Central-Kurdish Text-To-Speech (Sorani)

Dictionaries, terminologies and ontologies

Check out a comprehensive list of Kurdish dictionaries and beware of copyright issues in the following projects:

• Kurdish lexicographical resources in Ontolex-Lemon (Sorani, Kurmanji, Gorani and Southern Kurdish)
• Check Dolan Hêriş's repositories for a list of Kurdish dictionaries and tools to extract words
• KurdNet-the Kurdish wordNet (Sorani)
• Kurdish annotated lexicon (Sorani)
• Freedict word lists (Sorani and Kurmanji)
• Translation Initiative for COVID-19 including Sorani and Kurmanji
• MyMemory dictionaries with an open-access API (Sorani)

Datasets

• Manchester Database of Kurdish Dialects
• Dataset of Kurdish poems with meter and form tags
• A Twitter dataset (Sorani and Kurmanji)
• Datasets for text to Kurdish Sign Language (Sorani)
• A dataset for speech recognition (Sorani)
• Universal dependency (Kurmanji)
• Web Inventory of Transcribed and Translated Talks (WIT3) (Sorani)
• Sorani and Kurmanji morphological datasets in UniMorph
• FakeKurdNews, an annotated dataset for Sorani Kurdish fake news detection
• profanity language (Sorani)

Benchmarks

• Morphological analysis:
  • KurdishHunspell evaluation datasets (Sorani)
• Tokenization:
  • KurdishTokenization (Sorani, Kurmanji)
  • A sentence-segmented dataset (Sorani)
• Transliteration
  • Wergor transliteration datasets
  • Evaluation datasets for Kurdish Grapheme-to-Phoneme Conversion systems (Sorani)
• Spelling error correction
  • Central-Kurdish-Spelling-dataset

Other resources

Word Embeddings:

• fastText word vectors (Sorani and Kurmanji)
• Polyglot's word embeddings

Tools

Fundamental processing

• Language identifier (Sorani and Kurmanji)
• Wergor for transliteration (Sorani and Kurmanji)
• Kurdish Tokenization
• Jedar stemmer
• Apertium project for Kurmanji and Sorani morphological analysis
• Kurdish Hunspell for Sorani morphological analysig, spell checking, stemming and lemmatization
• A finite-state morphological analyzer for Central Kurdish (Sorani)
• Part-of-speech tagger (Sorani)
• Alexina Framework: morphological analysis and POS-tagger for Sorani (soralex) and Kurmanji (kurlex)
• Kurdspell for Sorani spell checking
• Apertium rule-based Sorani spell-checker
• Gende Stemmer (Sorani)
• Conversion of numbers into words (Sorani and Kurmanji)
• Conversion of words into IPA (Kurmanji)

Machine translation

• Apertium (Sorani and Kurmanji)
• Kurdish MT (Sorani)

Named-entity recognition

• Autoregressive Entity Retrieval (Kurmanji)

Optical character recognition

• Kurdish Handwritten Words (Sorani)

Libraries

• Kurdish Language Processing Toolkit: a natural language processing toolkit in Python
• Kurdînûs: pure JavaScript tools for transliteration, text conversion and normalization
• Kurdish Language Library: converting characters and digits in Persian, English and Arabic to Kurdish and vice versa
• AsoSoft's Library for Kurdish: normalizer, numeral converter, grapheme-to-phoneme convertor in C#

Other

In addition to these, you can find further information in other repositories and pages as follows:

• Developers Tree
• Kurdish resources

Research

These references are provided based on the data collected in the paper entitled KLPT – Kurdish Language Processing Toolkit. Note that references are provided in the bibliography file.

Reference	Year	Field	dialects
esmaili2013sorani	2013	Dialectology	Sorani, Kurmanji
hassani2016automatic	2016	Dialectology	Sorani, Kurmanji
malmasi2016subdialectal	2016	Dialectology	Sorani
al2017kurdish	2017	Dialectology	Sorani, Kurmanji, Gorani
amani:hal-03262435	2021	Dialectology	Kurdish, Zazaki & Gorani
mohammed2012automatic	2012	Information retrieval and Text mining	Sorani
esmaili2012challenges	2012	Information retrieval and Text mining	Sorani
littell2016named	2016	Information retrieval and Text mining	Sorani
hassani2017method	2017	Information retrieval and Text mining	Sorani, Kurmanji
esmaAl-Talabaniili2014towards	2014	Information retrieval and Text mining	Sorani, Kurmanji
jaf2016simple	2016	Information retrieval and Text mining	Sorani
rashid2017robust	2017	Information retrieval and Text mining	Sorani
rashid2017automatic	2017	Information retrieval and Text mining	Sorani
saeed2018improving	2018	Information retrieval and Text mining	Sorani
mustafa2018kurdish	2018	Information retrieval and Text mining	Sorani
saeed2018evaluation	2018	Information retrieval and Text mining	Sorani
ahmadi2019wergor	2019	Information retrieval and Text mining	Sorani
mahmudi2021automated	2021	Information retrieval and Text mining	Sorani
abdulrahman2022lmspell	2022	Information retrieval and Text mining	Sorani
esmaili2013building	2013	Lexical resources	Sorani
aliabadi2014towards	2014	Lexical resources	Sorani
aliabadi2014semi	2014	Lexical resources	Sorani
ataman2018bianet	2018	Lexical resources	Kurmanji
ahmadi2019towards	2019	Lexical resources	Sorani, Kurmanji, Gorani
abdulrahman2019developing	2019	Lexical resources	Sorani
abdulrahman2020using	2020	Lexical resources	Sorani
veisi2020toward	2020	Lexical resources	Sorani
ahmadi2020corpus	2020	Lexical resources	Sorani
ahmadi-2020-building	2020	Lexical resources	Zaza, Gorani
veisi2021jira	2021	Lexical resources	Sorani
azin2021sk	2021	Lexical resources	Southern Kurdish
hassani2017kurdish	2017	Machine Translation	Sorani, Kurmanji
kaka2018english	2018	Machine Translation	Sorani
ahmadi2020machine	2020	Machine Translation	Sorani
goyal2021flores	2021	Machine Translation	101 languages incl. Sorani
amini2021central	2021	Machine Translation	Sorani
ahmadi2022leveraging	2022	Machine Translation	Sorani
baban1995programmable	1995	Morphological and syntactic analysis	Sorani
walther2010developing	2010	Morphological and syntactic analysis	Sorani
walther2010fast	2010	Morphological and syntactic analysis	Kurmanji
salavati2013stemming	2013	Morphological and syntactic analysis	Sorani
jaf2014stemmer	2014	Morphological and syntactic analysis	Sorani
jaf2016chapter	2016	Morphological and syntactic analysis	Sorani
gokirmak2017dependency	2017	Morphological and syntactic analysis	Kurmanji
salavati2018building	2018	Morphological and syntactic analysis	Sorani
mustafa2018kurdish	2018	Morphological and syntactic analysis	Sorani
ahmadi2020towards	2020	Morphological and syntactic analysis	Sorani
ahmadi-2020-tokenization	2020	Morphological and syntactic analysis	Sorani, Kurmanji
ahmadi2021modelling	2021	Morphological and syntactic analysis	Sorani
ahmadi2020Hunspell	2021	Morphological and syntactic analysis	Sorani
naserzade2021ckmorph	2021	Morphological and syntactic analysis	Sorani
mohammed2012uniqueness	2012	Optical character recognition	Sorani
mohammed2013handwritten	2013	Optical character recognition	Sorani
shaltookisentiment	2016	Optical character recognition	Sorani
zarro2017recognition	2017	Optical character recognition	Sorani
yaseen2018kurdish	2018	Optical character recognition	Sorani
dinler2018kurdish	2018	Optical character recognition	Sorani
app11209752	2021	Optical character recognition	Sorani
kaka2017building	2017	Other	Sorani
mahmudi2021automatic	2021	Other	Sorani
ahmadi2021ickl	2021	Other	Sorani
hashim2018kurdish	2018	Sign language recognition	Sorani
kamal-hassani-2020-towards	2020	Sign language recognition	Sorani
daneshfar2009implementation	2009	Speech recognition	Sorani
barkhoda2009comparison	2009	Speech recognition	Sorani
bahrampour2009implementation	2009	Speech recognition	Sorani
hassani2011kurdish	2011	Speech recognition	Sorani
dinler2017formant	2017	Speech recognition	Kurmanji
dinler2018extraction	2018	Speech recognition	Sorani, Kurmanji
qader2019kurdish	2019	Speech recognition	Sorani
ahmadi-2020-klpt	2020	Toolkits	Sorani, Kurmanji
de2021multilingual	2021	Named-entity recognition	Kurmanji
abdullah2022	2022	Sentiment analysis	Sorani
awlla2022	2022	Sentiment analysis	Sorani
amin2022kurdish	2022	Sentiment analysis	Sorani
zuhair2021	2021	Other	Sorani
kamala2022kurdish	2022	Other	Sorani

Cite this repository

If you find the provided data useful for your project, feel free to use it and please, cite the following paper, too:

@inproceedings{ahmadi-2020-klpt,
    title = "{KLPT} {--} {K}urdish Language Processing Toolkit",
    author = "Ahmadi, Sina",
    booktitle = "Proceedings of Second Workshop for NLP Open Source Software (NLP-OSS)",
    month = nov,
    year = "2020",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://www.aclweb.org/anthology/2020.nlposs-1.11",
    doi = "10.18653/v1/2020.nlposs-1.11",
    pages = "72--84"
}