Home / Query / WordAlign / Wiki     [books] [DGT] [DOGC] [ECB] [EMEA] [EUbooks] [EU] [Europarl] [giga] [GNOME] [GlobalVoices] [hren] [JRC] [KDE4/doc] [MBS] [memat] [MontenegrinSubs] [MultiUN] [NCv9/v11] [OO/OO3] [subs/16/18] [ParaCrawl] [ParCor] [PHP] [SETIMES] [SPC] [Tatoeba] [TEP] [TedTalks] [TED] [Tanzil] [Ubuntu] [UN] [WikiSource] [Wikipedia] [WMT] [XhosaNavy]

PHP

A parallel corpus originally extracted from http://se.php.net/download-docs.php. The original documents are written in English and have been partly translated into 21 languages. The original manuals contain about 500,000 words. The amount of actually translated texts varies for different languages between 50,000 and 380,000 words. The corpus is rather noisy and may include parts from the English original in some of the translations. The corpus is tokenized and each language pair has been sentence aligned.

23 languages, 252 bitexts
total number of files: 71,414
total number of tokens: 3.28M
total number of sentence fragments: 1.38M

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

cs de en es fi fr he hu it ja ko nl pl pt_BR ro ru sk sl sv tr tw zh zh_TW
cs viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewcs
de ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewde
en ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewen
es ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewes
fi ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewfi
fr ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewfr
he ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewviewhe
hu ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewviewhu
it ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewviewit
ja ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewviewja
ko ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewviewko
nl ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewviewnl
pl ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewviewpl
pt_BR ces ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewviewpt_BR
ro ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewviewro
ru ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewviewru
sk ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewviewsk
sl ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewviewsl
sv ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewviewsv
tr ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewviewviewtr
tw ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewtw
zh ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces viewzh
zh_TW ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces ces zh_TW
cs de en es fi fr he hu it ja ko nl pl pt_BR ro ru sk sl sv tr tw zh zh_TW

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 sentencescs de en es fi fr he hu it ja ko nl pl pt_BR ro ru sk sl sv tr twzh zh_TW
cs 3,271 0.1M 63.5k5.4k 5.8k 4.2k 1.1k 5.4k 2.0k 3.6k 3.7k 5.5k 5.1k 3.2k 3.7k 4.1k 1.9k 1.8k 2.2k 2.0k 1.6k 2.7k 6.1k 5.4k
de 3,268 0.3M 65.1k32.1k 12.3k 7.1k 1.3k 11.0k 3.0k 6.7k 5.9k 8.4k 7.8k 4.2k 6.6k 6.4k 2.9k 3.0k 3.0k 2.9k 2.5k 5.0k 10.4k 9.8k
en 3,283 0.5M 66.8k33.0k 39.7k 8.5k 0.9k 16.0k 3.9k 7.8k 8.0k 12.0k 10.9k 4.4k 7.7k 8.1k 3.7k 3.7k 4.3k 3.7k 3.3k 9.1k 18.4k 18.3k
es 3,231 0.2M 59.6k30.3k 33.2k 34.9k 1.3k 8.4k 2.5k 4.8k 4.5k 6.9k 6.4k 3.3k 4.1k 4.5k 2.4k 2.3k 2.6k 2.4k 2.1k 3.7k 7.8k 7.2k
fi 3,283 57.6k 66.8k27.4k 27.4k 27.9k 26.7k 1.7k 1.1k 1.2k 1.1k 3.1k 3.0k 1.0k 1.5k 1.4k 1.3k 1.4k 1.3k 1.3k 1.2k 1.4k 3.1k 2.2k
fr 3,241 0.4M 60.8k30.8k 36.8k 42.2k 33.6k 26.6k 3.4k 7.5k 6.8k 10.1k 9.3k 4.4k 6.7k 6.7k 3.5k 3.4k 3.9k 3.5k 3.1k 5.8k 12.5k 12.3k
he 3,071 0.1M 58.1k25.3k 26.2k 27.8k 25.2k 24.5k 25.9k 2.3k 2.1k 4.5k 4.0k 1.6k 2.7k 2.6k 1.7k 1.8k 1.7k 1.8k 1.6k 3.9k 4.4k 3.9k
hu 3,279 0.2M 66.1k30.3k 33.8k 35.4k 31.0k 27.8k 33.3k 25.7k 4.8k 6.2k 5.7k 3.3k 5.6k 4.9k 2.5k 2.4k 2.3k 2.4k 1.9k 4.0k 7.2k 6.6k
it 3,273 0.2M 65.1k30.5k 32.6k 35.5k 30.5k 27.7k 32.3k 25.6k 31.7k 6.3k 5.9k 3.1k 4.3k 4.5k 2.4k 2.1k 2.3k 2.2k 1.9k 3.0k 7.5k 6.9k
ja 3,277 0 18.2k8.9k 11.7k 15.5k 10.1k 6.5k 13.2k 7.6k 9.6k 9.7k 9.7k 4.7k 5.3k 5.4k 3.6k 3.8k 3.8k 3.5k 3.4k 4.8k 8.6k 11.1k
ko 3,238 0 16.3k8.5k 11.1k 14.3k 9.7k 6.2k 12.3k 7.0k 8.9k 9.1k 13.7k 4.4k 4.8k 4.8k 3.4k 3.4k 3.5k 3.3k 3.2k 4.4k 8.2k 10.2k
nl 3,277 0.1M 66.4k30.0k 30.8k 32.0k 29.2k 27.9k 29.7k 25.0k 30.3k 30.1k 8.1k 7.6k 3.5k 3.9k 1.9k 1.8k 1.9k 1.8k 1.6k 2.5k 5.2k 4.6k
pl 3,280 0.2M 73.7k29.8k 32.8k 34.6k 29.5k 27.9k 31.8k 25.9k 32.5k 30.7k 6.6k 6.2k 29.9k 7.9k 5.1k 5.0k 5.2k 5.2k 4.8k 5.9k 7.8k 7.9k
pt_BR 3,283 0.2M 73.5k30.2k 32.5k 35.3k 29.9k 28.0k 31.5k 25.8k 31.4k 30.9k 6.8k 6.1k 30.5k 39.0k 5.0k 5.1k 5.4k 5.1k 4.8k 5.2k 8.2k 8.2k
ro 3,264 93.8k 72.7k27.7k 28.9k 30.4k 27.7k 27.7k 28.0k 24.7k 28.8k 28.5k 5.0k 4.7k 28.2k 35.5k 35.5k 4.7k 4.7k 4.5k 4.6k 4.4k 6.1k 5.7k
ru 3,283 0.1M 72.4k27.4k 28.3k 30.1k 27.0k 27.6k 27.7k 25.0k 28.1k 27.8k 5.4k 4.9k 27.8k 34.8k 35.0k 34.3k 4.8k 4.7k 4.7k 4.4k 5.6k 5.8k
sk 3,285 93.7k 74.1k28.1k 28.4k 31.2k 27.6k 27.8k 28.4k 24.8k 28.4k 28.4k 5.2k 4.8k 28.2k 35.5k 36.0k 34.9k 34.4k 4.6k 4.6k 4.2k 6.0k 5.9k
sl 3,276 88.9k 73.6k27.9k 28.7k 30.6k 27.5k 27.9k 28.1k 25.0k 28.6k 28.4k 4.9k 4.5k 28.2k 35.9k 36.0k 35.1k 34.6k 35.0k 4.6k 4.4k 5.8k 5.6k
sv 3,285 76.2k 74.1k27.7k 28.4k 30.2k 27.3k 27.9k 27.7k 24.7k 28.2k 28.1k 4.8k 4.5k 28.1k 35.5k 35.7k 35.2k 34.7k 35.2k 35.4k 4.2k 5.6k 5.4k
tr 2,904 0.2M 61.6k24.9k 27.7k 32.7k 25.5k 23.8k 27.5k 26.5k 26.7k 25.5k 7.4k 6.9k 24.8k 31.7k 30.7k 29.8k 29.0k 29.5k 29.9k 29.6k 7.4k 6.8k
tw0 0 0
zh 3,277 0 61.8k27.3k 31.6k 41.7k 28.7k 24.5k 33.4k 24.3k 28.4k 28.9k 12.2k 11.8k 26.6k 31.7k 32.0k 29.7k 29.6k 29.5k 29.7k 29.4k 28.5k 45.0k
zh_TW 3,285 0 68.6k30.6k 35.2k 46.5k 31.9k 27.7k 36.9k 26.7k 32.0k 32.3k 15.0k 14.0k 30.1k 36.7k 37.1k 34.3k 34.7k 34.5k 34.5k 34.3k 32.6k 16.7k

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.