amh: inflection tables for Amharic (ISO 639-3amh)amh.args: a UniMorph 4.0 compatible verion ofamhamh.gloss: English glosses for the lemmas inamhREADME.md: this file
All data were generated using the Amharic morphological generator in HornMorpho, described in Gasser (2013).
Decisions about lemma frequency and subcategorization were based on Biniam Gebremichael's Amharic word list.
Common orthographic variants of lemmas and common variant plural forms of nouns are included.
In these cases, the variants are distinguished with the features LGSPEC1 and LGSPEC2.
Predictable orthographic variants are not included.
These include alternate representations of labialized consonants (ቆ/ቈ, ቁ/ቍ, ፏ/ፍዋ, etc.), the two ways of writing the vowel ɨ/i before y (e.g., ጅያ/ጂያ), and the common replacement of ው with ዉ.
The total number of lemmas is 2,465: V 670, N 1599, ADJ 196.
Although the formation of the passive-reflexive, causative, reciprocal, and frequentative from bare roots is relatively productive, many roots fail to appear in the basic, underived form, in which case the morphologically passive-reflexive or causative form is not semantically or syntactically passive-reflexive or causative (for example, "passive-reflexive" ተቀመጠ 'sit', "causative" አለቀሰ 'cry'). For this reason, each of these derived forms of a root is treated as a separate lemma, and only those appearing in the word list have been included. For example, based on the root 'break', there are the following lemmas in the repository: ሰበረ 'break', ተሰበረ 'be broken', ሰባበረ 'break repeatedly'. Other possible forms derived from this root are not included.
Verb lemmas appear in all of the major tense-aspect categories and the infinitive:
- perfective (UM:
PFV) - finite (UM:
IPFV) and non-finite imperfective (UM:NFIN;IPFV) - gerundive (UM:
V.CVB) - present perfect (UM:
PRF) - jussive-imperative (UM:
IMP) - infinitive (UM:
V.MSDR).
Features for the copula (ነው) and verb of existence (ለአ) include "present" (PRS).
Verb lemmas are divided into two subcategories based on their occurrence with different subjects or with only (or mainly) third person singular masculine subjects. Lemmas in the first ("personal") category are further divided into reciprocal verbs, which only appear with plural subjects, and all other verbs, which appear with all eight subject agreement affixes. Lemmas in the "impersonal" category only appear with third person singular masculine affixes.
Not included in the paradigms: object suffixes, negative prefixes/suffixes, relative prefixes, prepositional/conjunctive prefixes.
Nouns and adjectives share most of their morphology and are often not clearly distinguished.
Noun and adjective lemmas are divided into five categories based on their occurrence with the masculine and feminine definite suffixes, with the plural form, and with possessive suffixes. Those classified as adjectives admit both masculine and feminine definite suffixes and plural forms and have English adjective glosses. Since most nouns do have inherent gender, gender is only specified for singular nouns with one of the gender-specific definite suffixes. Gender is not distinguished for plural nouns and adjectives.
Not included in the paradigms: prepositional/case prefixes and the accusative suffix.
Michael Gasser
ቃል /k'al/ 'word'
ቃል ቃሉ N;DEF;MASC
ቃል ቃሉ N;PSS3SM
ቃል ቃላቱ N;DEF;PL
ቃል ቃላቱ N;PL;PSS3SM
ቃል ቃላታቸው N;PL;PSS3P
ቃል ቃላታችሁ N;PL;PSS2P
ቃል ቃላታችን N;PL;PSS1P
ቃል ቃላቴ N;PL;PSS1S
ቃል ቃላት N;PL
ቃል ቃላትህ N;PL;PSS2SM
ቃል ቃላትሽ N;PL;PSS2SF
ቃል ቃላትዎ N;PL;FORM;PSS2S
ቃል ቃላቷ N;PL;PSS3SF
ቃል ቃላቸው N;PSS3P
ቃል ቃላችሁ N;PSS2P
ቃል ቃላችን N;PSS1P
ቃል ቃሌ N;PSS1S
ቃል ቃል N
ቃል ቃልህ N;PSS2SM
ቃል ቃልሽ N;PSS2SF
ቃል ቃልዎ N;FORM;PSS2S
ቃል ቃሎቹ N;DEF;PL;LGSPEC1
ቃል ቃሎቹ N;PL;PSS3SM;LGSPEC1
ቃል ቃሎቻቸው N;PL;PSS3P;LGSPEC1
ቃል ቃሎቻችሁ N;PL;PSS2P;LGSPEC1
ቃል ቃሎቻችን N;PL;PSS1P;LGSPEC1
ቃል ቃሎቼ N;PL;PSS1S;LGSPEC1
ቃል ቃሎች N;PL;LGSPEC1
ቃል ቃሎችህ N;PL;PSS2SM;LGSPEC1
ቃል ቃሎችሽ N;PL;PSS2SF;LGSPEC1
ቃል ቃሎችዎ N;PL;FORM;PSS2S;LGSPEC1
ቃል ቃሎቿ N;PL;PSS3SF;LGSPEC1
ቃል ቃሏ N;DEF;FEM
ቃል ቃሏ N;PSS3SF
ተቀመጠ /tǝk'ǝmmǝt'ǝ/ 'sit'
ተቀመጠ ተቀምጠን V.CVB;1;PL
ተቀመጠ ተቀምጬ V.CVB;1;SG
ተቀመጠ ተቀምጣችሁ V.CVB;2;PL
ተቀመጠ ተቀምጠሽ V.CVB;2;SG;FEM
ተቀመጠ ተቀምጠህ V.CVB;2;SG;MASC
ተቀመጠ ተቀምጠው V.CVB;3;PL
ተቀመጠ ተቀምጣ V.CVB;3;SG;FEM
ተቀመጠ ተቀምጦ V.CVB;3;SG;MASC
ተቀመጠ መቀመጥ V.MSDR
ተቀመጠ እንቀመጥ V;IMP;1;PL
ተቀመጠ ልቀመጥ V;IMP;1;SG
ተቀመጠ ተቀመጡ V;IMP;2;PL
ተቀመጠ ተቀመጪ V;IMP;2;SG;FEM
ተቀመጠ ተቀመጥ V;IMP;2;SG;MASC
ተቀመጠ ይቀመጡ V;IMP;3;PL
ተቀመጠ ትቀመጥ V;IMP;3;SG;FEM
ተቀመጠ ይቀመጥ V;IMP;3;SG;MASC
ተቀመጠ እንቀመጣለን V;IPFV;1;PL
ተቀመጠ እቀመጣለሁ V;IPFV;1;SG
ተቀመጠ ትቀመጣላችሁ V;IPFV;2;PL
ተቀመጠ ትቀመጫለሽ V;IPFV;2;SG;FEM
ተቀመጠ ትቀመጣለህ V;IPFV;2;SG;MASC
ተቀመጠ ይቀመጣሉ V;IPFV;3;PL
ተቀመጠ ትቀመጣለች V;IPFV;3;SG;FEM
ተቀመጠ ይቀመጣል V;IPFV;3;SG;MASC
ተቀመጠ እንቀመጥ V;IPFV;NFIN;1;PL
ተቀመጠ እቀመጥ V;IPFV;NFIN;1;SG
ተቀመጠ ትቀመጡ V;IPFV;NFIN;2;PL
ተቀመጠ ትቀመጪ V;IPFV;NFIN;2;SG;FEM
ተቀመጠ ትቀመጥ V;IPFV;NFIN;2;SG;MASC
ተቀመጠ ይቀመጡ V;IPFV;NFIN;3;PL
ተቀመጠ ትቀመጥ V;IPFV;NFIN;3;SG;FEM
ተቀመጠ ይቀመጥ V;IPFV;NFIN;3;SG;MASC
ተቀመጠ ተቀመጥን V;PFV;1;PL
ተቀመጠ ተቀመጥኩ V;PFV;1;SG
ተቀመጠ ተቀመጣችሁ V;PFV;2;PL
ተቀመጠ ተቀመጥሽ V;PFV;2;SG;FEM
ተቀመጠ ተቀመጥክ V;PFV;2;SG;MASC
ተቀመጠ ተቀመጡ V;PFV;3;PL
ተቀመጠ ተቀመጠች V;PFV;3;SG;FEM
ተቀመጠ ተቀመጠ V;PFV;3;SG;MASC
ተቀመጠ ተቀምጠናል V;PRF;1;PL
ተቀመጠ ተቀምጫለሁ V;PRF;1;SG
ተቀመጠ ተቀምጣችኋል V;PRF;2;PL
ተቀመጠ ተቀምጠሻል V;PRF;2;SG;FEM
ተቀመጠ ተቀምጠሀል V;PRF;2;SG;MASC
ተቀመጠ ተቀምጠዋል V;PRF;3;PL
ተቀመጠ ተቀምጣለች V;PRF;3;SG;FEM
ተቀመጠ ተቀምጧል V;PRF;3;SG;MASC
@inproceedings{pimentel-ryskina-etal-2021-sigmorphon,
title = "SIGMORPHON 2021 Shared Task on Morphological Reinflection: Generalization Across Languages",
author = "Pimentel, Tiago and
Ryskina, Maria and
Mielke, Sabrina J. and
Wu, Shijie and
Chodroff, Eleanor and
Leonard, Brian and
Nicolai, Garrett and
Ghanggo Ate, Yustinus and
Khalifa, Salam and
Habash, Nizar and
El-Khaissi, Charbel and
Goldman, Omer and
Gasser, Michael and
Lane, William and
Coler, Matt and
Oncevay, Arturo and
Montoya Samame, Jaime Rafael and
Silva Villegas, Gema Celeste and
Ek, Adam and
Bernardy, Jean-Philippe and
Shcherbakov, Andrey and
Bayyr-ool, Aziyana and
Sheifer, Karina and
Ganieva, Sofya and
Plugaryov, Matvey and
Klyachko, Elena and
Salehi, Ali and
Krizhanovsky, Andrew and
Krizhanovsky, Natalia and
Vania, Clara and
Ivanova, Sardana and
Salchak, Aelita and
Straughn, Christopher and
Liu, Zoey and
Washington, Jonathan North and
Ataman, Duygu and
Kiera{\'s}, Witold and
Woli{\'n}ski, Marcin and
Suhardijanto, Totok and
Stoehr, Niklas and
Nuriah, Zahroh and
Ratan, Shyam and
Tyers, Francis M. and
Ponti, Edoardo M. and
Aiton, Grant and
Hatcher, Richard J. and
Prud'hommeaux, Emily and
Kumar, Ritesh and
Hulden, Mans and
Barta, Botond and
Lakatos, Dorina and
Szolnok, G{\'a}bor and
{\'A}cs, Judit and
Raj, Mohit and
Yarowsky, David and
Cotterell, Ryan and
Ambridge, Ben and
Vylomova, Ekaterina",
booktitle = "Proceedings of the 18th SIGMORPHON Workshop on Computational Research in Phonetics, Phonology, and Morphology",
month = aug,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.sigmorphon-1.25",
doi = "10.18653/v1/2021.sigmorphon-1.25",
pages = "229--259",
abstract = "This year's iteration of the SIGMORPHON Shared Task on morphological reinflection focuses on typological diversity and cross-lingual variation of morphosyntactic features. In terms of the task, we enrich UniMorph with new data for 32 languages from 13 language families, with most of them being under-resourced: Kunwinjku, Classical Syriac, Arabic (Modern Standard, Egyptian, Gulf), Hebrew, Amharic, Aymara, Magahi, Braj, Kurdish (Central, Northern, Southern), Polish, Karelian, Livvi, Ludic, Veps, V{\~o}ro, Evenki, Xibe, Tuvan, Sakha, Turkish, Indonesian, Kodi, Seneca, Ash{\'a}ninka, Yanesha, Chukchi, Itelmen, Eibela. We evaluate six systems on the new data and conduct an extensive error analysis of the systems' predictions. Transformer-based models generally demonstrate superior performance on the majority of languages, achieving {\textgreater}90{\%} accuracy on 65{\%} of them. The languages on which systems yielded low accuracy are mainly under-resourced, with a limited amount of data. Most errors made by the systems are due to allomorphy, honorificity, and form variation. In addition, we observe that systems especially struggle to inflect multiword lemmas. The systems also produce misspelled forms or end up in repetitive loops (e.g., RNN-based models). Finally, we report a large drop in systems' performance on previously unseen lemmas.",
}
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent