Research on Modern Latvian Language and
Development of Language Technology



The National Corpus Collection (NCC) has been created, incorporating more than 10 Latvian language corpora developed by various institutions. It is now accessible through the platform, which features a unified search system.

The 27th International Scientific Conference "Word and Its Aspects of Research" was held at Liepaja University, where several presentations were delivered by researchers involved in the LATE project. Daiga Straupeniece and Elza Ozola (LiepU) presented on "The Development Process of the Latvian Sign Language Corpus", Guna Rābante-Buša (LU MII) on "Variations in the Pronunciation of Particles in Different Word Combinations", and Ieva Auziņa (LU LVI) on "The Connection between Some Prefix and Adverb Forms with Non-Derivative and Derivative Verbs".

Ilze Auziņa and Baiba Saulīte presented a paper on the National Corpus Collection and its potential uses at the conference "Latvian Language in the European Union - Language Technology in Public Administration and Society".

At the International conference, "Human Language Technologies - Baltic Perspective" (Baltic HLT) held in Riga, Sanita Reinsone (LU LFMI) and Matīss Rikters (University of Tokyo) delivered a presentation on "How Masterly Are People at Playing with Their Vocabulary?". The paper was published in the Scopus-indexed Journal of Modern Baltic Computing. 

The practical seminar "Digital Resources for Linguists" featured a presentation by Baiba Saulīte and Ilze Auziņa (LU MII) on the use of corpora in linguistic research.

Andra Kalnača and Ilze Lokmane (LU HZF) presented on "Expressive Predicative Constructions in Latvian" at the conference "EXPRESSIVES. Theoretical and Experimental Approaches to the Expressive Content. 9th Experimental Pragmatics Conference 2022" organized by the University of Genoa.

At the 55th Annual Meeting of the Societas Linguistica Europaea held at the University of Bucharest in Romania, Andra Kalnača and Ilze Lokmane (LU HZF) presented on "Syntactic Constructions with the Verb Vajadzēt 'to Need, Must' in Latvian".

Emīlija Mežale (LU HZF) presented on "Article-like Usage of Demonstrative Pronouns: A Case of Finnish and Latvian" at the conference "Congressus XIII Internationalis Fenno-Ugristarum" organized by the University of Vienna.

Andra Kalnača, Daiga Degsne, and Tatjana Pakalne (LU HZF) presented on "Latvian Deverbal Nouns in -ien- and -um- and Derivational Productivity: A Corpus-based Analysis" at the conference "Grammar and Corpora" organized by Ghent University.
IMCS UL participates in the LREC 2022 conference
LREC 2022 (13th Conference on Language Resources and Evaluation) is taking place in Marseille on June 20-25, 2022. LREC is the major event on Language Resources and Evaluation for Human Language Technologies. The Institute of Mathematics and Computer Science, University of Latvia, participates in the conference with three presentations.
On June 21, there will be a report on the Latvian Language Learner Corpus LaVA, which includes more than 1000 essays written by language learners studying at Latvian universities (corpus size – 190k words). By analyzing the mistakes of the language learners marked in the texts, a set of self-assessment exercise was created.
On June 22, we will present a poster about extending the online dictionary with word sense synonyms and other semantic links, building a new lexico-semantic resource – Latvian WordNet.
Poster presentation on the Latvian National Corpora Collection (LNCC) - a diverse collection of corpora representing both written and spoken language is planned on Midsummer's Day, June 23. All corpora of LNCC are annotated with a uniform morpho-syntactic annotation scheme, enabling federated search and consistent linguistics analysis in more than 20 corpora (1.3B tokens).

Emīlija Mežale from LU HZF participated in the summer school "Speech Matters" in Como, Italy. The event focused on various issues related to the study of spoken language, including grammar, emphasis and gesture analysis, peculiarities of different community languages, reference, speech synthesis, and the basics of using the ELAN annotation tool.

Emīlija Mežale from LU HZF took part in the conference "Subjectivity and Intersubjectivity in Language and Culture", organized by Tartu University and the Estonian Literary Museum. She presented on the topic of "(Im)politeness of colloquial language features in Latvian and Finnish."

Ilze Lokmane from LU HZF and Madara Stāde from LU MII participated in the section "Language and Culture in the Digital Age" at the 80th International Scientific Conference of the University of Latvia. They presented a lecture titled "Latvian Lexical Network as a Novel Digital Language Processing Tool: Achievements and Prospects."

Andra Kalnača and Tatjana Pakalne from LU HZF participated in the 6th Nordic and Baltic Digital Humanities Conference (DHNB2022) at Uppsala University in Sweden. They presented on the topic of "Assigning Meaning to Novel Productively Formed Complex Words in Actual Language Use: A Case of the Latvian Agentive Suffix -tāj-."

The 80th International Scientific Conference of Latvian and General Linguistics was held at the University of Latvia took place. The conference featured a section on "Grammar and Corpus Studies" which was organized by project participants Andra Kalnača and Ilze Lokmane from the HZF LU. Participants from the project: Linda Lauze, Ilze Auziņa, Laura Rituma, Baiba Saulīte, Gunta Nespore-Bērzkalne, Lauma Pretkalniņa, Ilze Auziņa, Kristīne Levāne-Petrova, Roberts Darģis, Ilze Lokmane, Andra Kalnača, Tatjana Pakalne, and Emīlija Mežale. Altogether, the participants from the LATE project delivered eight presentations during the conference.

Ilze Migla from LU LVI took part in the "Baltic Languages: synchrony and diachrony" section of the international scientific conference "XXXII Scientific Readings." She presented on the topic of "Lexeme putns (bird) in Latvian phraseology."

LATE kick-off meeting 

On 18 January, the LATE project kick-off meeting took place on Zoom. It was attended by representatives from all partner institutions:
LU MII: Ilze Auziņa, Baiba Saulīte, Normunds Grūzītis, Inguna Skadiņa, Vita Matule
LU LVI: Edmunds Trumpa, Agris Timuška, Sanda Rapa, Marita Silkāne
LU HZF: Andra Kalnača, Ilze Lokmane
LU LFMI: Sanita Reinsone
LiepU: Dina Bethere, Airita Lindberga

Online hands-on workshop on platform and searching the corpora, organized by the Institute of Mathematics and Computer Sciences of the University of Latvia in cooperation with CLARIN Latvia. Instructors: Ilze Auziņa and Baiba Saulīte.

Information about the project

The project "Research on Modern Latvian Language and Development of Language Technology” is implemented within the framework of the National Research Programme "Letonika – Fostering a Latvian and European Society".

Project No: VPP-LETONIKA-2021/1-0006

Implementation period: 20.12.2021–19.12.2024

Project funding: 1 068 000 EUR

Funded by: Latvian Council of Science of the Ministry of Education and Science

Project partners: Institute of Mathematics and Computer Science of the University of Latvia (leading partner), University of Latvia (Latvian Language Institute of the University of Latvia and Faculty of Humanities of the University of Latvia), Institute of Literature, Folklore and Art of the University of Latvia, Liepāja University


Project leader: Ilze Auziņa

The aim of the project is to advance research on the grammatical, lexical-semantic, phonetic and phonological system of the modern Latvian language, and Latvian sign language using data-driven methods, as well as to develop sustainable Latvian language resources and tools. In order to achieve the goal, the Latvian speech corpus, the pilot corpus of Latvian sign language will be developed, and and “Dictionary of Contemporary Latvian” will be improved. Based on Latvian grammar studies, “Latvian Treebank” will be enhanced. These resources will be integrated into a single Latvian language research infrastructure, as well into the CLARIN-LV repository. During the project, a LATE platform for speech transcription and subtitling will be created.

Project's research group:  10 lead participants and 33 participants (including 14 student-participants).