Digital language resources in Oxford
- British National Corpus:
- IVIE Corpus of English dialects
- Oxford English Corpus, access available on demand to Oxford researchers - please apply via the website
- Sketch Engine - institutional access funded by ELEXIS started 1 April 2018
- Various Corpora from the OTA (Oxford Text Archive), hosted in Oxford since 1976
- Literary and linguistic electronic resources on SOLO. For further information contact Johanneke Sytsema
The University of Oxford has licences for 1997, 2008, 2009, 2010, 2013 and 2015 for the Linguistic Data Consortium. Take a look at their catalogue, and if there is something there that you are interested in, and you don't see it in the list below, please get in touch with Martin Wynne. Thanks to OUP who paid for the 2009 licence in full for the University, Department of Computer Science who paid for the 2010 and 2015 licences, and the Phonetics Laboratory for 1997 and 2013. The following resources have been downloaded from the LDC and are now available online from IT Services for Oxford users. Consult the LDC catalogue for the full list of what is available, and get in touch with ota at bodleian.ox.ac.uk. Please note that you are bound by the terms and conditions of the user agreements associated with each of these resources, which can be found on the LDC website.
- LDC93S1 TIMIT Acoustic-Phonetic Continuous Speech Corpus (436 Mb)
- LDC96L14 CELEX2 (71 Mb)
- LDC96S37 CALLHOME Japanese Speech (1.5 Gb)
- LDC96S53 CALLFRIEND Japanese (975 Mb)
- LDC97S62 Switchboard-1 Release 2 (10.3 Gb)
- LDC2002L49 Buckwalter Arabic Morphological Analyzer Version 1.0 (1.1 Mb)
- LDC2004L02 Buckwalter Arabic Morphological Analyzer Version 2.0 (1.1 Mb)
- LDC2008S09 CHAracterizing INdividual Speakers (CHAINS) (2.8 Gb)
- LDC2008T18 New York Times Annotated Corpus (3.3 Gb)
- LDC2009T03 GALE Phase 1 Arabic Newsgroup Parallel Text - Part 1 (8.5 Mb)
- LDC2009T09 GALE Phase 1 Arabic Newsgroup Parallel Text - Part 2 (2.4 Mb)
- LDC2009T10 Language Understanding Annotation Corpus (1.6 Mb)
- LDC2009T12 2008 CoNLL Shared Task Data (16.1 Mb)
- LDC2009T13 English Gigaword Fourth Edition disk 1 (4 Gb) and LDC2009T13 English Gigaword Fourth Edition disk 2 (4.5 Gb)
- LDC2009T22 Arabic Newswire English Translation Collection (4 Mb)
- LDC2009T23 FactBank 1.0 (4 Mb)
- LDC2009T24 OntoNotes Release 3.0 (448 Mb)
- LDC2009T30 Arabic Gigaword Fourth Edition (2.7 Gb)
- LDC2010L01 LDC Standard Arabic Morphological Analyzer (SAMA) Version 3.1 (1.2 Mb)
- LDC2013S06 LDC Spoken Language Sampler - Second Release (97 Mb)
- LDC2013T18 Semantic Textual Similarity (STS) 2013 Machine Translation (77 kb)
- LDC2013T19 OntoNotes Release 5.0 (932 Mb)
- LDC2015T13 English News Text Treebank: Penn Treebank Revised (9.7 Mb)
Please visit the LDC website for more information about these resources, and to consult the relevant licence agreements. Note that these resources are for use by members of the University of Oxford, and you are not permitted to redistribute them.
The UK is a member of the CLARIN European Research Infrastructure Consortium, which offers easy access to language data and tools for research in the humanities and social sciences. The latest up to date information on activities and resources can be found at CLARIN website. The University of Oxford is home to the co-ordination of the CLARIN-UK Consortium.
- CLARIN protected resources including resources in Czech, Danish, Dutch, English, German, Norwegian, and online interfaces to a number of other languages via the Corpuscle archive at the University of Bergen, including Abkhazian, Bulgarian, Older Scots, Persian, Slovenian, among others. In most cases, you need to log in to these sites by following the link to 'Log in via your institution' or 'EduGAIN', or simply 'Log in', and you will be redirected to WebAuth.
- Virtual Language Observatory is the gateway to a larger number of resources. The VLO is a resource discovery service aggregating records for resources held in most of the major archives world-wide.
- CLARIN Resource Showcases is a way to explore a small selection of open access online corpora and lexical resources offered by CLARIN, including resources in Czech, Dutch, Finnish, German and Swedish
- CLARIN-UK also makes available to users in Oxford a number of important resources, provided by the members of the CLARIN-UK Consortium.
- Oral History & Technology: a new website which will feature tools for processing audio data, including speech synthesis and alignment.
There are further corpora, copies of which may be available in Oxford, but under a variety of different licensing and access arrangements (often on optical disk). Please get in touch to add to the list. For these resources, contact Martin Wynne unless otherwise stated.
- BNC XML version, BNC Baby (sampler on one CD)
- Corpus of Spoken Dutch
- Corpus of Spoken Japanese
- IPI-PAN corpus of Polish
- COLT Corpus of London Teenagers' Speech
- Gesprochenes Jiddisch Textzeugen einer Europäisch-jüdischen Kultur
- ICAME corpus collection
- East meets West: a compendium of multilingual resources (the TELRI CD, parallel aligned corpora in many European languages)
- An informal interest group in corpus linguistics meets termly. Subscribe to the mailing list here
- A course in Corpus Linguistics is held at IT Services in Hilary Term each year, usually Thursdays 12:30 to 13:30 - more details here (look for session names starting with 'Corpus').
- Developing Linguistic Corpora: a free online guide to good practice in creating a corpus