Home / Query / WordAlign / Wiki    [ada83] [bible] [bianet] [books] [CAPES] [DGT] [DOGC] [ECB] [EhuHac] [Elhuyar] [EMEA] [EUbooks] [EU] [Europarl] [finlex] [fiskmö] [giga] [GNOME] [GlobalVoices] [hren] [infopankki] [JRC] [JW300] [KDE4/doc] [MBS] [memat] [MontenegrinSubs] [MultiUN] [MultiParaCrawl] [NCv9/v11/v14] [Ofis] [OO/OO3] [subs/16/18] [ParaCrawl] [ParCor] [PHP] [QED] [sardware] [SciELO] [SETIMES] [SPC] [Tatoeba] [Tanzil] [TEP] [TedTalks] [TED] [Tilde] [Ubuntu] [UN] [UNPC] [Wikimedia] [Wikipedia] [WikiSource] [WMT] [XhosaNavy]

MultiParaCrawl v5

Parallel corpora from Web Crawls collected in the ParaCrawl project and further processed for making it a multi-parallel corpus by pivoting via English. Here we only provide the additional language pairs that came out of pivoting. The bitexts for English are available from the ParaCrawl release.

23 languages, 253 bitexts
total number of files: 23
total number of tokens: 5.43G
total number of sentence fragments: 223.92M

Please, acknowledge the ParaCrawl project at http://paracrawl.eu. This version is derived from the original release at their website adjusted for redistribution via the OPUS corpus collection. Please, acknowledge OPUS as well for this service.

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.

The data is released under the Creative Commons CC0 license ("no rights reserved"), see http://paracrawl.eu/download.html for more details

Release history:

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 cs da de el es et fi fr ga hr hu it lt lv mt nl pl pt ro sk sl sv
bg viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewbg
cs ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewcs
da ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewda
de ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewde
el ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewel
es ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewes
et ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewet
fi ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewfi
fr ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewfr
ga ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewga
hr ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewhr
hu ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewhu
it ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewit
lt ces ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewlt
lv ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewlv
mt ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewmt
nl ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewnl
pl ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewpl
pt ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewpt
ro ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewro
sk ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewviewsk
sl ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewsl
sv ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces sv
bg cs da de el es et fi fr ga hr hu it lt lv mt nl pl pt ro sk sl sv

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 sentencesbg cs da de el es et fi fr ga hr hu it lt lv mt nl pl pt ro sk sl sv
bg 1 57.7M 2.6M0.5M 0.4M 0.7M 0.4M 0.7M 0.3M 0.4M 0.8M 96.6k 0.3M 0.3M 0.5M 0.3M 0.2M 68.0k 0.4M 0.4M 0.5M 0.4M 0.4M 0.2M 0.4M
cs 1 119.0M 5.3M0.5M 0.8M 1.4M 0.6M 1.2M 0.4M 0.6M 1.3M 0.1M 0.4M 0.6M 1.2M 0.3M 0.3M 79.1k 0.9M 1.0M 1.0M 0.6M 0.8M 0.3M 0.7M
da 1 108.3M 4.7M0.4M 0.8M 1.4M 0.6M 1.4M 0.4M 0.8M 1.4M 0.1M 0.4M 0.5M 1.3M 0.3M 0.3M 88.3k 1.3M 0.9M 1.2M 0.5M 0.5M 0.3M 1.3M
de 1 909.7M 38.3M0.7M 1.4M 1.4M 0.8M 7.0M 0.4M 0.8M 8.1M 0.1M 0.5M 0.8M 6.0M 0.4M 0.3M 82.8k 3.1M 1.8M 3.6M 0.7M 0.6M 0.4M 1.4M
el 1 94.9M 3.8M0.4M 0.6M 0.6M 0.8M 1.0M 0.2M 0.5M 1.0M 0.1M 0.3M 0.4M 0.9M 0.3M 0.2M 76.1k 0.7M 0.6M 0.9M 0.5M 0.4M 0.3M 0.6M
es 1 961.5M 38.7M0.7M 1.3M 1.5M 7.1M 1.0M 0.4M 0.9M 9.9M 0.1M 0.5M 0.8M 6.8M 0.4M 0.3M 78.2k 2.9M 1.8M 6.0M 0.9M 0.6M 0.3M 1.4M
et 1 26.5M 1.4M0.3M 0.4M 0.4M 0.4M 0.2M 0.4M 0.4M 0.4M 95.1k 0.2M 0.3M 0.3M 0.3M 0.2M 81.2k 0.3M 0.3M 0.3M 0.3M 0.2M 0.2M 0.4M
fi 1 54.4M 3.2M0.4M 0.6M 0.8M 0.8M 0.5M 0.9M 0.4M 1.0M 0.1M 0.3M 0.5M 0.8M 0.3M 0.3M 80.7k 0.8M 0.7M 0.8M 0.5M 0.4M 0.2M 1.2M
fr 1 1.3G 51.1M0.8M 1.4M 1.4M 8.3M 1.0M 10.1M 0.4M 1.0M 0.1M 0.5M 0.8M 7.1M 0.4M 0.3M 82.3k 3.4M 1.8M 4.6M 0.9M 0.6M 0.4M 1.4M
ga 1 24.8M 0.8M97.6k 0.1M 0.1M 0.1M 0.1M 0.1M 96.4k 0.1M 0.1M 67.5k 0.1M 0.1M 78.0k 75.7k 54.7k 99.4k 98.2k 0.1M 75.6k 0.1M 76.7k 96.5k
hr 1 43.2M 1.9M0.3M 0.5M 0.4M 0.5M 0.3M 0.5M 0.2M 0.3M 0.5M 68.2k 0.3M 0.6M 0.2M 0.2M 50.6k 0.4M 0.4M 0.4M 0.3M 0.3M 0.3M 0.3M
hu 1 107.0M 4.1M0.3M 0.7M 0.5M 0.8M 0.4M 0.8M 0.3M 0.5M 0.9M 0.1M 0.3M 0.8M 0.3M 0.3M 76.4k 0.6M 0.7M 0.6M 0.6M 0.5M 0.3M 0.5M
it 1 562.3M 22.0M0.5M 1.2M 1.3M 6.1M 1.0M 7.0M 0.4M 0.8M 7.2M 0.1M 0.6M 0.8M 0.4M 0.3M 91.4k 2.6M 1.7M 3.9M 0.9M 0.6M 0.4M 1.3M
lt 1 25.6M 1.3M0.3M 0.3M 0.3M 0.4M 0.3M 0.4M 0.3M 0.4M 0.4M 79.0k 0.2M 0.3M 0.4M 0.3M 73.4k 0.4M 0.4M 0.4M 0.3M 0.3M 0.2M 0.4M
lv 1 22.5M 1.1M0.2M 0.3M 0.3M 0.3M 0.2M 0.3M 0.2M 0.3M 0.3M 76.5k 0.2M 0.3M 0.3M 0.3M 66.9k 0.3M 0.3M 0.3M 0.2M 0.3M 0.2M 0.3M
mt 1 4.2M 0.2M68.4k 79.5k 88.8k 83.3k 76.5k 78.7k 81.7k 81.1k 82.9k 55.0k 50.8k 76.8k 92.0k 73.8k 67.2k 85.7k 86.2k 87.2k 68.7k 82.6k 71.5k 86.5k
nl 1 237.9M 10.6M0.4M 0.9M 1.3M 3.1M 0.8M 3.0M 0.3M 0.8M 3.5M 0.1M 0.4M 0.6M 2.7M 0.4M 0.3M 86.4k 1.2M 2.2M 0.6M 0.5M 0.3M 1.3M
pl 1 144.8M 6.7M0.4M 1.1M 0.9M 1.9M 0.6M 1.9M 0.3M 0.7M 1.8M 99.3k 0.4M 0.7M 1.8M 0.4M 0.3M 86.8k 1.2M 1.5M 0.7M 0.6M 0.3M 1.0M
pt 1 320.1M 13.5M0.5M 1.0M 1.2M 3.6M 0.9M 6.1M 0.3M 0.8M 4.7M 0.1M 0.4M 0.7M 4.0M 0.4M 0.3M 87.9k 2.2M 1.6M 0.7M 0.5M 0.3M 1.2M
ro 1 65.7M 2.9M0.4M 0.6M 0.5M 0.7M 0.5M 0.9M 0.3M 0.5M 0.9M 76.4k 0.3M 0.6M 0.9M 0.3M 0.2M 69.2k 0.6M 0.7M 0.7M 0.4M 0.3M 0.6M
sk 1 41.6M 2.1M0.4M 0.8M 0.5M 0.6M 0.4M 0.6M 0.2M 0.4M 0.6M 0.1M 0.3M 0.5M 0.6M 0.3M 0.3M 83.1k 0.5M 0.6M 0.5M 0.4M 0.4M 0.5M
sl 1 31.8M 1.5M0.2M 0.3M 0.3M 0.4M 0.3M 0.3M 0.2M 0.2M 0.4M 77.5k 0.3M 0.3M 0.4M 0.2M 0.2M 71.9k 0.3M 0.4M 0.3M 0.3M 0.4M 0.3M
sv 1 131.5M 6.1M0.4M 0.7M 1.3M 1.4M 0.6M 1.5M 0.4M 1.2M 1.4M 97.5k 0.3M 0.5M 1.4M 0.4M 0.3M 87.1k 1.3M 1.0M 1.2M 0.6M 0.5M 0.3M

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.