SETIMES

A parallel corpus of news articles in the Balkan languages, originally extracted from http://www.setimes.com. The corpus is compiled by Nikola Ljubešić and is taken from http://www.nljubesic.net/resources/corpora/setimes provided under the CC-BY-SA license

10 languages, 45 bitexts
total number of files: 90
total number of tokens: 425.89M
total number of sentence fragments: 17.60M

Please cite the following article if you use any part of the corpus in your own work:
J. Tiedemann, 2012, Parallel Data, Tools and Interfaces in OPUS. In Proceedings of the 8th International Conference on Language Resources and Evaluation (LREC 2012)

Download

Below you can download data files for all language pairs in different formats and with different kind of annotation (if available). You can click on the various links as explained below. In addition to the files shown on this webpage, OPUS also provides pre-compiled word alignments and phrase tables, bilingual dictionaries, frequency counts, and these files can be found through the resources search form on the top-level website of OPUS.

Bottom-left triangle: download files

ces = sentence alignments in XCES format
leftmost column language IDs = tokenized corpus files in XML
TMX and plain text files (Moses): see "Statistics" below
lower row language IDs = parsed corpus files (if they exist)

Upper-right triangle: sample files

view = bilingual XML file samples
upper row language IDs = monolingual XML file samples
rightmost column language IDs = untokenized corpus files

	bg	bs	el	en	hr	mk	ro	sq	sr	tr
bg		view	view	view	view	view	view	view	view	view	bg
bs	ces		view	view	view	view	view	view	view	view	bs
el	ces	ces		view	view	view	view	view	view	view	el
en	ces	ces	ces		view	view	view	view	view	view	en
hr	ces	ces	ces	ces		view	view	view	view	view	hr
mk	ces	ces	ces	ces	ces		view	view	view	view	mk
ro	ces	ces	ces	ces	ces	ces		view	view	view	ro
sq	ces	ces	ces	ces	ces	ces	ces		view	view	sq
sr	ces	ces	ces	ces	ces	ces	ces	ces		view	sr
tr	ces	ces	ces	ces	ces	ces	ces	ces	ces		tr
	bg	bs	el	en	hr	mk	ro	sq	sr	tr

Statistics and TMX/Moses Downloads

Number of files, tokens, and sentences per language (including non-parallel ones if they exist)
Number of sentence alignment units per language pair

Upper-right triangle: download translation memory files (TMX)
Bottom-left triangle: download plain text files (MOSES/GIZA++)
Language ID's, first row: monolingual plain text files (tokenized)
Language ID's, first column: monolingual plain text files (untokenized)


language	files	tokens	sentences	bg	bs	el	en	hr	mk	ro	sq	sr	tr
bg	9	44.6M	1.8M		0.1M	0.2M	0.2M	0.2M	0.2M	0.2M	0.2M	0.2M	0.2M
bs	9	28.3M	1.2M	0.1M		0.1M	0.1M	0.1M	0.1M	0.1M	0.1M	0.1M	0.1M
el	9	47.4M	1.9M	0.2M	0.1M		0.2M	0.2M	0.2M	0.2M	0.2M	0.2M	0.2M
en	9	44.8M	1.9M	0.2M	0.1M	0.2M		0.2M	0.2M	0.2M	0.2M	0.2M	0.2M
hr	9	40.2M	1.8M	0.2M	0.1M	0.2M	0.2M		0.2M	0.2M	0.2M	0.2M	0.2M
mk	9	44.6M	1.8M	0.2M	0.1M	0.2M	0.2M	0.2M		0.2M	0.2M	0.2M	0.2M
ro	9	46.8M	1.8M	0.2M	0.1M	0.2M	0.2M	0.2M	0.2M		0.2M	0.2M	0.2M
sq	9	48.2M	1.9M	0.2M	0.1M	0.2M	0.2M	0.2M	0.2M	0.2M		0.2M	0.2M
sr	9	42.9M	1.8M	0.2M	0.1M	0.2M	0.2M	0.2M	0.2M	0.2M	0.2M		0.2M
tr	9	38.3M	1.8M	0.2M	0.1M	0.2M	0.2M	0.2M	0.2M	0.2M	0.2M	0.2M

Note that TMX files only contain unique translation units and, therefore, the number of aligned units is smaller than for the distributions in Moses and XML format. Moses downloads include all non-empty alignment units including duplicates. Token counts for each language also include duplicate sentences and documents.

Disclaimer

We do not own any of the text from which the data has been extracted.
We only offer files that we believe we are free to redistribute. If any doubt occurs about the legality of any of our file downloads we will take them off right away after contacting us.

Notice and take down policy

Notice: Should you consider that our data contains material that is owned by you and should therefore not be reproduced here, please:

Clearly identify yourself, with detailed contact data such as an address, telephone number or email address at which you can be contacted.
Clearly identify the copyrighted work claimed to be infringed.
Clearly identify the material that is claimed to be infringing and information reasonably sufficient to allow us to locate the material.
And contact the OPUS project at the following email address: opus-project at helsinki.fi.

Take down: We will comply to legitimate requests by removing the affected sources from the next release of the corpus.