Difference between revisions of "Turkic languages"

From Apertium
Jump to navigation Jump to search
Line 92: Line 92:
| <code>[[apertium-kaa]]</code> || [[Karakalpak]] || <code>-</code> || <code>kaa</code> || HFST (lexc+twol) || development ||align="right"| {{#lst:apertium-kaa/stats|stems}}||align="center"| [[apertium-kaa#Current_State|~{{:apertium-kaa/stats/average}}%]] || [[apertium-kaa]]&nbsp;([[incubator]]) || [[User:Francis Tyers|Fran]]
| <code>[[apertium-kaa]]</code> || [[Karakalpak]] || <code>-</code> || <code>kaa</code> || HFST (lexc+twol) || development ||align="right"| {{#lst:apertium-kaa/stats|stems}}||align="center"| [[apertium-kaa#Current_State|~{{:apertium-kaa/stats/average}}%]] || [[apertium-kaa]]&nbsp;([[incubator]]) || [[User:Francis Tyers|Fran]]
| <code>[[apertium-tyv]]</code> || [[Tuvan]] || <code>-</code> || <code>tuv</code> || HFST (lexc+twol) || development ||align="right"| {{#lst:apertium-tuv/stats|stems}}||align="center"| [[apertium-tuv#Current_State|~{{:apertium-tuv/stats/average}}%]] || [[apertium-tuv]]&nbsp;([[incubator]]) || [[User:Francis Tyers|Fran]]
| <code>[[apertium-aze]]</code> || [[Azerbaijani]] || <code>az</code> || <code>aze</code> || SFST || not known to work || || || [[apertium-tur-aze]] ([[staging]]) || [[User:zfe|Gianluca]]
| <code>[[apertium-aze]]</code> || [[Azerbaijani]] || <code>az</code> || <code>aze</code> || SFST || not known to work || || || [[apertium-tur-aze]] ([[staging]]) || [[User:zfe|Gianluca]]

Revision as of 13:35, 30 October 2013

The Turkic languages include Turkish, Azeri, Uzbek, Uyghur, Kazakh, Kyrgyz, Tatar, Chuvash, Turkmen, Bashkir, Sakha, and several dozen other languages. The languages are related with varying levels of mutual intelligibility. A team called Apertium Turkic aims to enhance MT coverage of these languages with apertium.

The ultimate master plan involves generating independent finite-state transducers for each language, and then making individual dictionaries and transfer rules for every pair. The current status of these goals is listed below.


The ultimate goal is to have multi-purposable transducers for a variety of Turkic languages. These can then be paired for X→Y translation with the addition of a CG for language X and transfer rules / dictionary for the pair X→Y. Below is listed development progress for each language's transducers and dictionary pairs.


See also: Turkic lexicon

Once a transducer has ~80% coverage on a range of medium-large corpora we can say it is "working". Over 90% and it can be considered to be "production".

name Language ISO 639 formalism state stems coverage location primary authors
-2 -3
apertium-kaz Kazakh kk kaz HFST (lexc+twol) production 36,595 ~94.5% apertium-kaz (languages) Ilnar, Jonathan, Fran, Nathan
apertium-tat Tatar tt tat HFST (lexc+twol) working 55,702 ~91% apertium-tat (languages) Ilnar, Fran, Jonathan, Milli
apertium-kir Kyrgyz ky kir HFST (lexc+twol) working 8,564 ~90.4% apertium-kir (languages) Jonathan, Mirlan, Fran
apertium-tur Turkish tr tur HFST (lexc+twol) working 17,221 ~87.3% apertium-tur (incubator) Fran, Gianluca
apertium-chv Chuvash cv chv HFST (lexc+twol) development 8,579 ~85% apertium-chv (languages) Hèctor
apertium-kum Kumyk - kum HFST (lexc+twol) development 4,918 ~90.2% apertium-kum (languages) Fran, Jonathan
apertium-uzb Uzbek uz uzb HFST (lexc+twol) development 34,470 ~82.9% apertium-uzb (languages)
apertium-bak Bashkir ba bak HFST (lexc+twol) development 2,827 ~66% apertium-bak (languages) Fran, Jonathan, Ilnar, Milli
apertium-nog Nogay - nog HFST (lexc+twol) development 1,385 ~81.4% apertium-nog (languages) Fran, Jonathan
apertium-tuk Turkmen tk tuk HFST (lexc+twol) development 2,988 ~70.7% apertium-tuk (languages) Fran
apertium-kaa Karakalpak - kaa HFST (lexc+twol) development 25,545 ~81.4% apertium-kaa (incubator) Fran
apertium-tyv Tuvan - tuv HFST (lexc+twol) development Apertium-tuv/stats [[apertium-tuv#Current_State|~Apertium-tuv/stats/average%]] apertium-tuv (incubator) Fran
apertium-aze Azerbaijani az aze SFST not known to work apertium-tur-aze (staging) Gianluca


Some Turkic languages that are particularly similar to one another (and hence have high levels of mutual intelligibility) include those in the following list:

Chuvash is very distant from other Turkic languages and is not even partially mutually intelligible with any of them.

Table of dix progress

As counted with https://svn.code.sf.net/p/apertium/svn/trunk/apertium-tools/get_stems.py

tur aze tuk uzb kir kaz tat chv bak uig khk eng rus

Turkic-Turkic pairs

See also: Turkic-Turkic translator

Text in italic denotes language pairs under development / in the incubator. Regular text denotes a functioning language pair in staging, while text in bold denotes a stable well-working language pair in trunk.

tur aze tuk uzb kir kaz tat chv bak uig
tur tur-aze tur-tuk tur-uzb tr-ky tr-tt tr-cv
aze aze-tur
tuk tuk-tur
uzb tur-uzb
kir ky-tr ky-kk
kaz kaz-kir kaz-tat
tat tt-tr tat-kaz tt-ba
chv cv-tr
bak ba-tt

Pairs with non-Turkic languages

tur aze tuk uzb kir kaz tat chv bak uig
eng tr-en ky-en kaz-eng
ru cv-ru
mng/khk mn-kk


  • Stable release of apertium-kaz
  • Stable release of apertium-tat
  • Stable release of apertium-kaz-tat
  • Rework apertium-kir to match new standards
  • Bring apertium-bak up to date (based on apertium-tat)
  • Expand apertium-tat-bak
  • Beta release of apertium-kaz-kir
  • Expand apertium-tuk
  • Expand apertium-chv
  • Basic transducers for:
    • Khakas
    • Tuvan
    • Sakha
    • Shor
    • Qaralpaq (based on Tatar Kazakh, probably, no?)
    • Uzbek
    • Uyghur
    • Nogay
    • Kumyk

Getting involved

We have a work plan for developing Turkic-Turkic translators and are working on a how-to for building a Turkic lexicon. Please come talk to us on IRC or contact us on the apertium-turkic mailing list.


Rough guide to tagsets in various Turkic language transducers, with an eye to keeping stuff that is basically the same tagged the same. In the following table, A stands for Apertium and T stands for TRmorph (See also the general tagset list).

Phenomenon Morphology Description Tag(s) Language(s) Notes
Part of speech
Noun <n>
Proper noun <np>
Determiner <det>
Numeral <num>
Adjective <adj> incl. var/yok
Adverb <adv>
Pronoun <prn>
Verb <v>
Auxiliary verb <vaux>
Copula <cop>
Adverb <adv>
Postadverb <postadv>
Postposition <post>
Particle[1] <part>
Coordinating conjunction <cnjcoo>
Subordinating conjunction <cnjsub>
Adverbial conjunction <cnjadv>
Abbreviation <abbr>
Personal Title <title>
Interjection <ij> Әлбетте(kaz), жок(kir) (cf. Adj)
Proper noun types
Toponym <top>
Anthroponym <ant>
Patronym <pat>
Cognomen (Surname) <cog>
Acronym <acr>
Other <al>
Pronoun types
Personal <pers>
Ordinal <ord>
Demonstrative <dem>
Indefinite <ind>
Interrogative <itg>
Reflexive <ref>
Quantifier <qnt>
Positive <pst>
Negative <neg>
Comparative <comp>
Reciprocal <recip>
First person <p1>
Second person <p2>
Third person <p3>
Numeral types
Substantive Substantive form of numerals (when they are used as the head of the noun phrase) <subst>
Ordinal <ord> Chuvash: -мĂш
Distributive <dist> Chuvash: -шĂр
Collective <coll> Chuvash: -Ăн
Nominative case (unmarked) <nom>
Genitive case <gen>
Dative case <dat>
Locative case <loc>
Ablative case -DAn Case indicating movement away <abl> Pan-turkic
Comitative case <com>
Terminative case <ter> Chuvash: -ччен
Benefactive (Purposive) case <ben> Chuvash: -шĂн
Allative (Directive) case Case indicating motion towards something <all> Chuvash: -АллА
1st pers sg <px1sg>
1st pers pl <px1pl>
2nd pers sg <px2sg>
2nd pers pl <px2pl>
3rd pers sg <px3sg>
3rd pers pl <px3pl>
3rd pers sg or pl <px3sp>
Masculine <m>
Feminine <f>
Masculine / feminine <mf>
Singular <sg>
Plural <pl>
Tense, aspect, mood
Present tense <pres>
Present continuous tense <cont> Turkish: -{bI}yor
Evidential tenseless/past tense <evid> Turkish: -m{I}ş (<past><evid>)
Past tense <past> Kyrgyz: -{G}{A}н
Definite past tense <ifi> Turkish: -{D}{I}
Imperfect <pii> Turkish: Aorist + -m{A}kt{A}
Past habitual tense <pih> Turkish: Aorist + -{D}{I}
Future tense <fut> Turkish: -{bY}{A}c{A}{k}
Imperative Mood for giving orders <imp>A, <t_imp>T Pan-turkic Turkish: -ø
Conditional <cond> Turkish: -s{A}
Aorist <aor> Turkish: -{A}r or -{bI}r
Optative <opt> Turkish: -{bY}{A}, Kirghiz: -мAк>чI
Obligative <oblig> Turkish: -m{A}l{I}
Potential <pot> Kirghiz: -чUдAй
Not-yet tense <notyet> Kirghiz: -E элек
Non-finite verb forms
Gerund makes verbs usable as nouns <ger>, <vn>?
Verbal adjective makes verbs usable as adjectives <vadj>
Participle makes verb a matrix verb usable auxiliaries and modals <part>, <vadv>?
??? makes verb usable as first of a dual-predicate construction ??
Infinitive citation form of verb and use in certain constructions <inf>?
Gerund #1 <ger1> Turkish: -m{A}
Gerund #2 <ger2> Turkish: -m{A}{K}
Gerund #3 <ger3> Turkish: -{D}{I}{k}
Gerund #4 <ger4> Turkish: -{bY}{I}ş
Gerund #5 <ger5> Turkish: -{bY}{A}n
Gerund #6 <ger6> Turkish: -{bY}{A}r{A}k
Gerund #7 <ger10> Turkish: -{bY}{I}p
Future gerund #1 <fger> Turkish: -{bY}{A}c{A}{k}
Imperfect participle #1 <fger> Turkish: -{bY}{A}r{A}{k}
Productive verbal derivation
Passive <pass>
Causative <caus>
Cooperative <coop> -{I}ш(kir), -{I}с(kaz)
Transitive, переходный <tv>
Intransitive, непереходный <iv>
Modal/question/etc. "particles"
Question used with yes/no, focus, etc. question morphemes <qst> most-all {М}{А}(kaz), {B}{I}(kir), m{I}(tur); +ше(kaz), ч{I}(kir)
Emphatic used with imperative/optative and other coercive verb forms <emph> most +ш{I}(kaz), +ч{I}(kir),(tat), s{A}n{A}(tur)

Official poem

  • Kovayla bira içerim, ama sen bilmezsin. Yarın gelir misin?
  • Vedrəyle pivə içirəm, ama sen bilməzsən. Yarın gələrmisən?
  • Чиләк белән сыра эчәм, әмма син белмисең. Иртәгә киләсеңме?
  • Шелекпен сыра ішемін, бірақ сен білмейсің. Ертең келесің бе?
  • Чака менен сыра ичем, бирок сен билбейсиң. Эртең келесиңби?
  • Челек булан пиво ичемен амма сен билмейсен. Эртен гелемисен?


  1. Warning: The use of the particle tag is highly discouraged.