The paper describes work-in-progress by the Pite Saami, Kola Saami and Izhva Komi language documentation projects, all of which record new spoken language data, digitize available recordings and annotate these multimedia data in order to provide comprehensive language corpora as databases for future research on and for endangered – and under-described – Uralic speech communities. Applying language technology in language documentation helps us to create more systematically annotated corpora, rather than eclectic data collections. Specifically, we describe a script providing interactivity between different morphosyntactic analysis modules implemented as Finite State Transducers and ELAN, a Graphical User Interface tool for annotating and presenting multimodal corpora. Ultimately, the spoken corpora created in our projects will be useful for scientifically significant quantitative investigations on these languages in the future.

The paper describes work-in-progress by the Pite Saami, Kola Saami and Izhva Komi language documentation projects, all of which use similar data and technical frameworks and are carried out collaboratively in Uppsala, Tromsø, Syktyvkar and Freiburg. Our projects record and annotate spoken language data in order to provide comprehensive speech corpora as databases for future research on and for these endangered-and under-described-Uralic speech communities. Applying language technology in language documentation helps us to create more systematically annotated corpora, rather than eclectic data collections. Ultimately, the multimodal corpora created by our projects will be useful for scientifically significant quantitative investigations on these languages in the future.

Language Science Press has no responsibility for the persistence or accuracy of URLs for external or third-party Internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate. Information regarding prices, travel timetables and other factual information given in this work are correct at the time of first publication but Language Science Press does not guarantee the accuracy of such information thereafter. Muv vienagijda Árjepluovest

25 AbstractThe paper describes work-in-progress by the Pite Saami, Kola Saami and Izhva Komi language documentation projects, all of which use similar data and technical frameworks and are carried out in Freiburg and in collaboration with Hamburg, Syktyvkar, Tromsø and Uppsala. Our projects work in the endangered language documentation framework and record new spoken language data, digitize available recordings and annotate these multimedia data in order to provide comprehensive language corpora as databases for future research on and for endangered and under-described Uralic speech communities. Applying NLP methods in language documentation -specifically rule-based morphological and syntactic analyzers -helps us to create more systematically annotated corpora, rather than eclectic data collections. We propose a step-by-step approach to reach higherlevel annotations by using and improving truly computational methods. Ultimately, the spoken corpora created by our projects will be useful for scientifically significant quantitative investigations on these languages in the future.

The main goal of this paper is to describe to what extent the three main open word classes in Pite Saami (nouns, verbs and adjectives) can be automatically assigned to inflectional classes in language technology, specifically for a Finite State Transducer. For each of these word classes, the relevant structural features necessary for determining inflectional class membership are described. In this, a clear difference between the behavior of nouns and verbs, on the one hand, and that of adjectives, on the other hand, is ascertained. While morphophonology, as seen in the paradigmatic behavior of all three word classes, is complex and features a number of types of stem alternations, nouns and verbs are predictable, while adjectives are not. With this in mind, a basic algorithm for extracting inflectional class assignment for nouns and verbs is presented for use in a LEXC framework. In contrast to this, adjectives must be assigned to inflectional classes manually. The main TWOLC rules used to trigger morphophonological alternations are also outlined. The Pite Saami lexicographic database that forms the backbone for the LEXC stem files is managed using FileMaker Pro database software, and the workflow used to extract and update LEXC files from that database is described, focussing on the differences between nouns and verbs, and adjectives. In this, light is shed on how, on the one hand, nominal and verbal inflectional patters are highly complex yet reliably systematic, while adjective morphophonology is complex and unpredictable. KokkuvõteSelle artikli peamine eesmärk on kirjeldada, mil määral saab kolme põhilist avatud sõnaklassi (substantiive, verbe ja adjektiive) pite saami keeles automaatselt flekteerida kasutades keeletehnoloogia FST-d. Artiklis kirjeldatakse iga sõ-naliigi muuttüübi määramiseks vajalikke strukturaalseid omadusi ning näidatak-se, et adjektiivid on substantiividest ja verbidest selgelt erinevad. Samal ajal kui kõigi kolme sõnaklassi paradigmaatilist käitumist iseloomustab kompleksne paljusid tüvevahelduse tüüpe hõlmav morfofonoloogia, saab substantiivide ja verbide muutumist ennustada, kuid adjektiivide oma mitte. Seega kirjeldatakse artiklis Tuuakse välja ka peamised TWOLC reeglid, mida kasutatakse morfofonoloogilise vahelduse tekitamiseks. LEXC tüvefailide põhialuseks on pite saami keele leksikograafiline andmebaas, mida hallatakse FileMaker tarkvaraga; artiklis kirjeldatakse sellest andmebaasist LEXC failide väljavõtmise ja nende uuendamise töö-voogu, keskendudes erinevustele nimisõnade ja verbide, ning adjektiivide vahel. Näidatakse, et substantiivide ja verbide fleksioonimustrid on küll komplekssed, kuid väga süstemaatilised, samas kui adjektiivide morfofonoloogia on keeruline ning raskesti ennustatav.