LAT service to be discontinued in November 2020


For technical reasons, the LAT service ( will be discontinued in the Language Bank of Finland as of November 30, 2020.

The LAT platform itself is no longer developed in its present form by MPI, and the media browser component Annex (part of LAT) is based on deprecated Adobe Flash technology that will stop working at the end of this year.

What will happen to the corpora that used to be available in LAT?

A replacement service for LAT has not yet been selected by the Language Bank of Finland. However, all the speech and sign language corpora that were previously available in LAT will be made available by alternative means.

All of the LAT corpora can be offered as downloadable packages that can be used and studied directly on the user’s local computer. In the download service, the corpora can be accessed under the same conditions as via LAT. For some corpora, more advanced solutions might already be available.

Which corpora are affected by LAT shutting down?

The LAT instances of the following corpora will be affected:

  • aku-egg: Speech and EGG (Electroglottography) Simultaneous Recordings
  • cfinsl-conv: Corpus of Finnish Sign Language: conversations
  • cfinsl-elicit: Corpus of Finnish Sign Language: elicited narratives
  • eduskunta-v1-lat: Plenary Sessions of the Parliament of Finland, Kielipankki LAT Version 1
  • elfa-lat: The Helsinki LAT Version of the ELFA Corpus
  • fbc-lat: The Helsinki LAT Version of the Finnish Broadcast Corpus
  • ffe: a single unpublished video file by an unknown creator, access restricted to the owner (this data will be archived temporarily but will be removed quite soon unless the owner turns up!)
  • finka: The Corpus of Border Karelia, Kielipankki LAT version
  • giellagas-north: Samples of Northern Saami
  • helpuhe1: The Longitudinal Corpus of Finnish Spoken in Helsinki (1970s, 1990s and 2010s)
  • kipo: The 2010 Language Policy Program of the Sign Languages of Finland Corpus (versions 1 and 2)
  • la-murre: The Finnish Dialect Syntax Archive’s Helsinki LAT Version
  • PeWi-corpus (the original authoritative copy is offered by MPI; the identical version will be removed from the Language Bank of Finland)
  • puheen-analyysi: Learning material for speech analysis
  • reittidemo-lat: The Helsinki LAT Version of the Route to A wing Corpus
  • seuruu: Follow-up Study of Dialects of Finnish
  • skn-lat: The Helsinki LAT Version of Samples of Spoken Finnish
  • snowfrog: ProGram data. The stories Snowman and Frog, where are you?
  • ssdc-2016: Skolt Saami Documentation Corpus (2016)

In order to see where each corpus will be located in the future, please refer to the corresponding metadata page where this information will be updated. The relevant metadata links are provided on the list above.

I came to this page via a PID, how do I know where the file is?

LAT assigned 25000 PIDs to individual files. We have no automatic mapping of these PIDs but we can help you find the file if you need it. We aimed to structure downloadable packages similar to the dataset structure on LAT. To locate a file look into the URL field of this page where you find a ”?path=...” parameter. Example: ?path=demo/TRASH/2017-01/526/v7556__.C_4.4_Viittomakielisten_kielelliset_oikeudet.imdi

This should help you locate your file, please contact us if you have any questions.


August-September 2020:

  • The (numerous) persistent identifiers assigned by LAT to individual files will be redirected to stopover/tombstone pages. There will be one tombstone page per dataset.
  • The LAT corpora that are not yet offered in Download will be moved there and their persistent identifiers will be updated to point to their new home.
  • Assuming that too many errors are not detected, this process should be complete by October.

September-October 2020:

  • Decisions are to be taken as to what kind of streaming services the Language Bank can implement for audio and video materials in the year 2021.

30th November 2020:

  • The support for the server where LAT is located will be discontinued and the service will be shut down.
  • All the corpora that were previously offered via LAT will continue to be offered at least for download.

Year 2021 (and later):

  • Given that there are sufficient resources, more functionalities for browsing, searching, and analyzing speech and sign language corpora can be added.

Further details on the schedule of the aforementioned process will be updated on this page. In case you need additional information at this point, please contact FIN-CLARIN directly.

Raja-Karjalan korpus (FINKA)

The Corpus of Border Karelia contains the audio recordings and transcripts of dialects spoken in the area of Border Karelia, where the very closely related varieties of eastern Finnish dialects and Karelian were in contact. The informants are evacuees who were mainly moved to eastern Finland after World War II.

The original interviews were recorded in the 1960s and the 1970s and transcribed at the University of Eastern Finland by various researchers using the Finno-Ugrian transcription system. The interviewees are elderly people who were born in the 1870s – 1910s.

The original material has been archived by the Institute for the Languages of Finland. During the FINKA project (funded by the Academy of Finland in 2011–2014), the transcripts were reviewed and reorganized into a machine-readable corpus that is compatible with modern research tools.

Viimeisimmät versiot: 
Raja-Karjalan korpus, Kielipankin LAT-versio (LAT-palvelussa; tämä versio poistuu marraskuussa 2020)
icon-info-circle Kuvailutiedot ja lisenssi
icon-quote-right Viittausohje tähän versioon
Avaa aineisto LAT-palvelussa icon-question-circle
Raja-Karjalan korpuksen ladattava versio
icon-info-circle Kuvailutiedot ja lisenssi
icon-quote-right Viittausohje tähän versioon
Lataa aineisto
Etsi muut saatavilla olevat versiot 

Raja-Karjalan korpus sisältää yhteensä noin 120 tuntia wav-muotoisia äänitiedostoja sekä niiden suomalais-ugrilaisella tarkekirjoituksella tuotetut transkriptiot sekä raakatekstimuodossa että ääneen kohdistettuina TextGrid-tiedostoina, joista on ”riisuttu” tarkemerkit. TextGrid-muotoisissa annotaatiotiedostoissa eri puhujien puheenvuorot on merkitty eri kerroksiin.

Litterointimerkkien kuvaus (pdf)

Äänitteet edustavat 1800-luvun lopulla ja 1900-luvun alussa syntyneiden rajakarjalaisten haastattelupuhetta, jota on tallennettu Kotimaisten kielten keskuksen Suomen kielen nauhoitearkiston kokoelmiin pääosin 1960- ja 1970-luvuilla. Näytteet edustavat Ilomantsin, Korpiselän, Suojärven, Suistamon, Impilahden ja Salmin pitäjien murretta.

Aineistosta on tarjolla toistaiseksi kaksi versiota. Aineistoa voi kuunnella ja litteraateista voi tehdä hakuja Kielipankin LAT-palvelussa. LAT-palvelusta voi myös ladata yksittäisten haastattelujen (”sessioiden”) tiedostoja niputettuina omalle koneelle. Aineistosta on myös saatavilla kokonaan omalle koneelle ladattava versio.

Tarkempaa tietoa eri aineistoversioiden sisällöstä löytyy niiden kuvailutiedoista.

Tietoa tämän aineiston LAT-version poistumisesta marraskuussa 2020

Kielipankin LAT-alusta poistuu käytöstä marraskuun 2020 aikana. Sen jälkeen tätä aineistoa ei enää pääse käyttämään LAT-näkymän kautta. Kaikki tähän aineistoon kuuluva LATissa ollut sisältö on kuitenkin saatavilla ladattavassa muodossa. Aineiston tutkimista ja käsittelyä voi siis edelleen jatkaa esimerkiksi ELAN-ohjelmalla.

Tämän aineistoryhmäsivun PID: