This small dataset contains 3 speech corpora collected using the Alex Translate telephone service (https://ufal.mff.cuni.cz/alex#alex-translate).
The "part1" and "part2" corpora contain English speech with transcriptions and Czech translations. These recordings were collected from users of the service. Part 1 contains earlier recordings, filtered to include only clean speech; Part 2 contains later recordings with no filtering applied.
The "cstest" corpus contains recordings of artificially created sentences, each containing one or more Czech names of places in the Czech Republic. These were recorded by a multinational group of students studying in Prague.
Human post-edited test sentences for the WMT 2017 Automatic post-editing task. This consists in 2,000 English sentences belonging to the IT domain and already tokenized. Source and target segments can be downloaded from: https://lindat.mff.cuni.cz/repository/xmlui/handle/11372/LRT-2132. All data is provided by the EU project QT21 (http://www.qt21.eu/).
Human post-edited test sentences for the WMT 2017 Automatic post-editing task. This consists in 2,000 German sentences belonging to the IT domain and already tokenized. Source and target segments can be downloaded from: https://lindat.mff.cuni.cz/repository/xmlui/handle/11372/LRT-2133. All data is provided by the EU project QT21 (http://www.qt21.eu/).
This dataset contains automatic paraphrases of Czech official reference translations for the Workshop on Statistical Machine Translation shared task. The data covers the years 2011, 2013 and 2014.
For each sentence, at most 10000 paraphrases were included (randomly selected from the full set).
The goal of using this dataset is to improve automatic evaluation of machine translation outputs.
If you use this work, please cite the following paper:
Tamchyna Aleš, Barančíková Petra: Automatic and Manual Paraphrases for MT Evaluation. In proceedings of LREC, 2016.
CUBBITT En-Cs translation models, exported via TensorFlow Serving, available in the Lindat translation service (https://lindat.mff.cuni.cz/services/translation/).
Models are compatible with Tensor2tensor version 1.6.6.
For details about the model training (data, model hyper-parameters), please contact the archive maintainer.
Evaluation on newstest2014 (BLEU):
en->cs: 27.6
cs->en: 34.4
(Evaluated using multeval: https://github.com/jhclark/multeval)
CUBBITT En-Fr translation models, exported via TensorFlow Serving, available in the Lindat translation service (https://lindat.mff.cuni.cz/services/translation/).
Models are compatible with Tensor2tensor version 1.6.6.
For details about the model training (data, model hyper-parameters), please contact the archive maintainer.
Evaluation on newstest2014 (BLEU):
en->fr: 38.2
fr->en: 36.7
(Evaluated using multeval: https://github.com/jhclark/multeval)
CUBBITT En-Pl translation models, exported via TensorFlow Serving, available in the Lindat translation service (https://lindat.mff.cuni.cz/services/translation/).
Models are compatible with Tensor2tensor version 1.6.6.
For details about the model training (data, model hyper-parameters), please contact the archive maintainer.
Evaluation on newstest2020 (BLEU):
en->pl: 12.3
pl->en: 20.0
(Evaluated using multeval: https://github.com/jhclark/multeval)
This submission contains trained end-to-end models for the Neural Monkey toolkit for Czech and English, solving three NLP tasks: machine translation, image captioning, and sentiment analysis.
The models are trained on standard datasets and achieve state-of-the-art or near state-of-the-art performance in the tasks.
The models are described in the accompanying paper.
The same models can also be invoked via the online demo: https://ufal.mff.cuni.cz/grants/lsd
There are several separate ZIP archives here, each containing one model solving one of the tasks for one language.
To use a model, you first need to install Neural Monkey: https://github.com/ufal/neuralmonkey
To ensure correct functioning of the model, please use the exact version of Neural Monkey specified by the commit hash stored in the 'git_commit' file in the model directory.
Each model directory contains a 'run.ini' Neural Monkey configuration file, to be used to run the model. See the Neural Monkey documentation to learn how to do that (you may need to update some paths to correspond to your filesystem organization).
The 'experiment.ini' file, which was used to train the model, is also included.
Then there are files containing the model itself, files containing the input and output vocabularies, etc.
For the sentiment analyzers, you should tokenize your input data using the Moses tokenizer: https://pypi.org/project/mosestokenizer/
For the machine translation, you do not need to tokenize the data, as this is done by the model.
For image captioning, you need to:
- download a trained ResNet: http://download.tensorflow.org/models/resnet_v2_50_2017_04_14.tar.gz
- clone the git repository with TensorFlow models: https://github.com/tensorflow/models
- preprocess the input images with the Neural Monkey 'scripts/imagenet_features.py' script (https://github.com/ufal/neuralmonkey/blob/master/scripts/imagenet_features.py) -- you need to specify the path to ResNet and to the TensorFlow models to this script
Feel free to contact the authors of this submission in case you run into problems!
This submission contains trained end-to-end models for the Neural Monkey toolkit for Czech and English, solving four NLP tasks: machine translation, image captioning, sentiment analysis, and summarization.
The models are trained on standard datasets and achieve state-of-the-art or near state-of-the-art performance in the tasks.
The models are described in the accompanying paper.
The same models can also be invoked via the online demo: https://ufal.mff.cuni.cz/grants/lsd
In addition to the models presented in the referenced paper (developed and published in 2018), we include models for automatic news summarization for Czech and English developed in 2019. The Czech models were trained using the SumeCzech dataset (https://www.aclweb.org/anthology/L18-1551.pdf), the English models were trained using the CNN-Daily Mail corpus (https://arxiv.org/pdf/1704.04368.pdf) using the standard recurrent sequence-to-sequence architecture.
There are several separate ZIP archives here, each containing one model solving one of the tasks for one language.
To use a model, you first need to install Neural Monkey: https://github.com/ufal/neuralmonkey
To ensure correct functioning of the model, please use the exact version of Neural Monkey specified by the commit hash stored in the 'git_commit' file in the model directory.
Each model directory contains a 'run.ini' Neural Monkey configuration file, to be used to run the model. See the Neural Monkey documentation to learn how to do that (you may need to update some paths to correspond to your filesystem organization).
The 'experiment.ini' file, which was used to train the model, is also included.
Then there are files containing the model itself, files containing the input and output vocabularies, etc.
For the sentiment analyzers, you should tokenize your input data using the Moses tokenizer: https://pypi.org/project/mosestokenizer/
For the machine translation, you do not need to tokenize the data, as this is done by the model.
For image captioning, you need to:
- download a trained ResNet: http://download.tensorflow.org/models/resnet_v2_50_2017_04_14.tar.gz
- clone the git repository with TensorFlow models: https://github.com/tensorflow/models
- preprocess the input images with the Neural Monkey 'scripts/imagenet_features.py' script (https://github.com/ufal/neuralmonkey/blob/master/scripts/imagenet_features.py) -- you need to specify the path to ResNet and to the TensorFlow models to this script
The summarization models require input that is tokenized with Moses Tokenizer (https://github.com/alvations/sacremoses) and lower-cased.
Feel free to contact the authors of this submission in case you run into problems!
The data set includes training, development and test data from the shared tasks on pronoun-focused machine translation and cross-lingual pronoun prediction from the EMNLP 2015 workshop on Discourse in Machine Translation (DiscoMT2015). The release also contains the submissions to the pronoun-focused machine translation along with the manual annotations used for the official evaluation as well as gold-standard annotations of pronoun coreference for the shared task test set.
English-Urdu parallel corpus is a collection of religious texts (Quran, Bible) in English and Urdu language with sentence alignments. The corpus can be used for experiments with statistical machine translation. Our modifications of crawled data include but are not limited to the following:
1- Manually corrected sentence alignment of the corpora.
2- Our data split (training-development-test) so that our published experiments can be reproduced.
3- Tokenization (optional, but needed to reproduce our experiments).
4- Normalization (optional) of e.g. European vs. Urdu numerals, European vs. Urdu punctuation, removal of Urdu diacritics.
This package contains an extended version of the test collection used in the CLEF eHealth Information Retrieval tasks in 2013--2015. Compared to the original version, it provides complete query translations into Czech, French, German, Hungarian, Polish, Spanish and Swedish and additional relevance assessment.
This machine translation test set contains 2223 Czech sentences collected within the FAUST project (https://ufal.mff.cuni.cz/grants/faust, http://hdl.handle.net/11234/1-3308).
Each original (noisy) sentence was normalized (clean1 and clean2) and translated to English independently by two translators.
Data
-------
Hausa Visual Genome 1.0, a multimodal dataset consisting of text and images suitable for English-to-Hausa multimodal machine translation tasks and multimodal research. We follow the same selection of short English segments (captions) and the associated images from Visual Genome as the dataset Hindi Visual Genome 1.1 has. We automatically translated the English captions to Hausa and manually post-edited, taking the associated images into account.
The training set contains 29K segments. Further 1K and 1.6K segments are provided in development and test sets, respectively, which follow the same (random) sampling from the original Hindi Visual Genome.
Additionally, a challenge test set of 1400 segments is available for the multi-modal task. This challenge test set was created in Hindi Visual Genome by searching for (particularly) ambiguous English words based on the embedding similarity and manually selecting those where the image helps to resolve the ambiguity.
Dataset Formats
-----------------------
The multimodal dataset contains both text and images.
The text parts of the dataset (train and test sets) are in simple tab-delimited plain text files.
All the text files have seven columns as follows:
Column1 - image_id
Column2 - X
Column3 - Y
Column4 - Width
Column5 - Height
Column6 - English Text
Column7 - Hausa Text
The image part contains the full images with the corresponding image_id as the file name. The X, Y, Width, and Height columns indicate the rectangular region in the image described by the caption.
Data Statistics
--------------------
The statistics of the current release are given below.
Parallel Corpus Statistics
-----------------------------------
Dataset Segments English Words Hausa Words
---------- -------- ------------- -----------
Train 28930 143106 140981
Dev 998 4922 4857
Test 1595 7853 7736
Challenge Test 1400 8186 8752
---------- -------- ------------- -----------
Total 32923 164067 162326
The word counts are approximate, prior to tokenization.
Citation
-----------
If you use this corpus, please cite the following paper:
@InProceedings{abdulmumin-EtAl:2022:LREC,
author = {Abdulmumin, Idris
and Dash, Satya Ranjan
and Dawud, Musa Abdullahi
and Parida, Shantipriya
and Muhammad, Shamsuddeen
and Ahmad, Ibrahim Sa'id
and Panda, Subhadarshi
and Bojar, Ond{\v{r}}ej
and Galadanci, Bashir Shehu
and Bello, Bello Shehu},
title = "{Hausa Visual Genome: A Dataset for Multi-Modal English to Hausa Machine Translation}",
booktitle = {Proceedings of the Language Resources and Evaluation Conference},
month = {June},
year = {2022},
address = {Marseille, France},
publisher = {European Language Resources Association},
pages = {6471--6479},
url = {https://aclanthology.org/2022.lrec-1.694}
}
Data
----
Hindi Visual Genome 1.0, a multimodal dataset consisting of text and images suitable for English-to-Hindi multimodal machine translation task and multimodal research. We have selected short English segments (captions) from Visual Genome along with associated images and automatically translated them to Hindi with manual post-editing, taking the associated images into account. The training set contains 29K segments. Further 1K and 1.6K segments are provided in a development and test sets, respectively, which follow the same (random) sampling from the original Hindi Visual Genome.
Additionally, a challenge test set of 1400 segments will be released for the WAT2019 multi-modal task. This challenge test set was created by searching for (particularly) ambiguous English words based on the embedding similarity and manually selecting those where the image helps to resolve the ambiguity.
Dataset Formats
--------------
The multimodal dataset contains both text and images.
The text parts of the dataset (train and test sets) are in simple tab-delimited plain text files.
All the text files have seven columns as follows:
Column1 - image_id
Column2 - X
Column3 - Y
Column4 - Width
Column5 - Height
Column6 - English Text
Column7 - Hindi Text
The image part contains the full images with the corresponding image_id as the file name. The X, Y, Width and Height columns indicate the rectangular region in the image described by the caption.
Data Statistics
----------------
The statistics of the current release is given below.
Parallel Corpus Statistics
---------------------------
Dataset Segments English Words Hindi Words
------- --------- ---------------- -------------
Train 28932 143178 136722
Dev 998 4922 4695
Test 1595 7852 7535
Challenge Test 1400 8185 8665 (Released separately)
------- --------- ---------------- -------------
Total 32925 164137 157617
The word counts are approximate, prior to tokenization.
Citation
--------
If you use this corpus, please cite the following paper:
@article{hindi-visual-genome:2019,
title={{Hindi Visual Genome: A Dataset for Multimodal English-to-Hindi Machine Translation}},
author={Parida, Shantipriya and Bojar, Ond{\v{r}}ej and Dash, Satya Ranjan},
journal={Computaci{\'o}n y Sistemas},
note={In print. Presented at CICLing 2019, La Rochelle, France},
year={2019},
}
This package contains data sets for development and testing of machine translation of medical search short queries between Czech, English, French, and German. The queries come from general public and medical experts. and This work was supported by the EU FP7 project Khresmoi (European Comission contract No. 257528). The language resources are distributed by the LINDAT/Clarin project of the Ministry of Education, Youth and Sports of the Czech Republic (project no. LM2010013).
We thank Health on the Net Foundation for granting the license for the English general public queries, TRIP database for granting the license for the English medical expert queries, and three anonymous translators and three medical experts for translating amd revising the data.
This package contains data sets for development and testing of machine translation of medical queries between Czech, English, French, German, Hungarian, Polish, Spanish ans Swedish. The queries come from general public and medical experts. This is version 2.0 extending the previous version by adding Hungarian, Polish, Spanish, and Swedish translations.
This package contains data sets for development and testing of machine translation of sentences from summaries of medical articles between Czech, English, French, and German. and This work was supported by the EU FP7 project Khresmoi (European Comission contract No. 257528). The language resources are distributed by the LINDAT/Clarin project of the Ministry of Education, Youth and Sports of the Czech Republic (project no. LM2010013). We thank all the data providers and copyright holders for providing the source data and anonymous experts for translating the sentences.
This package contains data sets for development (Section dev) and testing (Section test) of machine translation of sentences from summaries of medical articles between Czech, English, French, German, Hungarian, Polish, Spanish
and Swedish. Version 2.0 extends the previous version by adding Hungarian, Polish, Spanish, and Swedish translations.
"Large Scale Colloquial Persian Dataset" (LSCP) is hierarchically organized in asemantic taxonomy that focuses on multi-task informal Persian language understanding as a comprehensive problem. LSCP includes 120M sentences from 27M casual Persian tweets with its dependency relations in syntactic annotation, Part-of-speech tags, sentiment polarity and automatic translation of original Persian sentences in five different languages (EN, CS, DE, IT, HI).
Source code of the LINDAT Translation service frontend. The service provides a UI and a simple rest api that accesses machine translation models served by tensorflow serving.
The most recent version of the code is available at https://github.com/ufal/lindat_translation.
This toolkit comprises the tools and supporting scripts for unsupervised induction of dependency trees from raw texts or texts with already assigned part-of-speech tags. There are also scripts for simple machine translation based on unsupervised parsing and scripts for minimally supervised parsing into Universal-Dependencies style.
Document-level testsuite for evaluation of gender translation consistency.
Our Document-Level test set consists of selected English documents from the WMT21 newstest annotated with gender information. Czech unnanotated references are also added for convenience.
We semi-automatically annotated person names and pronouns to identify the gender of these elements as well as coreferences.
Our proposed annotation consists of three elements: (1) an ID, (2) an element class, and (3) gender.
The ID identifies a person's name and its occurrences (name and pronouns).
The element class identifies whether the tag refers to a name or a pronoun.
Finally, the gender information defines whether the element is masculine or feminine.
We performed a series of NLP techniques to automatically identify person names and coreferences.
This initial process resulted in a set containing 45 documents to be manually annotated.
Thus, we started a manual annotation of these documents to make sure they are correctly tagged.
See README.md for more details.
This data set contains four types of manual annotation of translation quality, focusing on the comparison of human and machine translation quality (aka human-parity). The machine translation system used is English-Czech CUNI Transformer (CUBBITT). The annotations distinguish adequacy, fluency and overall quality. One of the types is Translation Turing test - detecting whether the annotators can distinguish human from machine translation.
All the sentences are taken from the English-Czech test set newstest2018 (WMT2018 News translation shared task www.statmt.org/wmt18/translation-task.html), but only from the half with originally English sentences translated to Czech by a professional agency.
Manual classification of errors of Czech-Slovak translation according to the classification introduced by Vilar et al. [1]. First 50 sentences from WMT 2010 test set were translated by 5 MT systems (Česílko, Česílko2, Google Translate and two Moses setups) and MT errors were manually marked and classified. Classification was applied in MT systems comparison [3]. Reference translation is included.
References:
[1] David Vilar, Jia Xu, Luis Fernando D’Haro and Hermann Ney. Error Analysis of Machine Translation Output. In International Conference on Language Resources and Evaluation, pages 697-702. Genoa, Italy, May 2006.
[2] http://matrix.statmt.org/test_sets/list
[3] Ondřej Bojar, Petra Galuščáková, and Miroslav Týnovský. Evaluating Quality of Machine Translation from Czech to Slovak. In Markéta Lopatková, editor, Information Technologies - Applications and Theory, pages 3-9, September 2011 and This work has been supported by the grants Euro-MatrixPlus (FP7-ICT-2007-3-231720 of the EU and
7E09003 of the Czech Republic)
Manual classification of errors of English-Slovak translation according to the classification introduced by Vilar et al. [1]. 50 sentences randomly selected from WMT 2011 test set [2] were translated by 3 MT systems described in [3] and MT errors were manually marked and classified. Reference translation is included.
References:
[1] David Vilar, Jia Xu, Luis Fernando D’Haro and Hermann Ney. Error Analysis of Machine Translation Output. In International Conference on Language Resources and Evaluation, pages 697-702. Genoa, Italy, May 2006.
[2] http://www.statmt.org/wmt11/evaluation-task.html
[3] Petra Galuščáková and Ondřej Bojar. Improving SMT by Using Parallel Data of a Closely Related Language. In Human Language Technologies - The Baltic Perspective - Proceedings of the Fifth International Conference Baltic HLT 2012, volume 247 of Frontiers in AI and Applications, pages 58-65, Amsterdam, Netherlands, October 2012. IOS Press. and This work has been supported by the grant Euro-MatrixPlus (FP7-ICT-2007-3-231720 of the EU and
7E09003 of the Czech Republic)
Manually ranked outputs of Czech-Slovak translations. Three annotators manually ranked outputs of five MT systems (Česílko, Česílko2, Google Translate and two Moses setups) on three data sets (100 sentences randomly selected from books, 100 sentences randomly selected from Acquis corpus and 50 first sentences from WMT 2010 test set). Ranking was applied in MT systems comparison in [1].
References:
[1] Ondřej Bojar, Petra Galuščáková, and Miroslav Týnovský. Evaluating Quality of Machine Translation from Czech to Slovak. In Markéta Lopatková, editor, Information Technologies - Applications and Theory, pages 3-9, September 2011 and This work has been supported by the grant Euro-MatrixPlus (FP7-ICT-2007-3-231720 of the EU and
7E09003 of the Czech Republic)
En-De translation models, exported via TensorFlow Serving, available in the Lindat translation service (https://lindat.mff.cuni.cz/services/translation/).
The models were trained using the MCSQ social surveys dataset (available at https://repo.clarino.uib.no/xmlui/bitstream/handle/11509/142/mcsq_v3.zip).
Their main use should be in-domain translation of social surveys.
Models are compatible with Tensor2tensor version 1.6.6.
For details about the model training (data, model hyper-parameters), please contact the archive maintainer.
Evaluation on MCSQ test set (BLEU):
en->de: 67.5 (train: genuine in-domain MCSQ data only)
de->en: 75.0 (train: additional in-domain backtranslated MCSQ data)
(Evaluated using multeval: https://github.com/jhclark/multeval)
En-Ru translation models, exported via TensorFlow Serving, available in the Lindat translation service (https://lindat.mff.cuni.cz/services/translation/).
The models were trained using the MCSQ social surveys dataset (available at https://repo.clarino.uib.no/xmlui/bitstream/handle/11509/142/mcsq_v3.zip).
Their main use should be in-domain translation of social surveys.
Models are compatible with Tensor2tensor version 1.6.6.
For details about the model training (data, model hyper-parameters), please contact the archive maintainer.
Evaluation on MCSQ test set (BLEU):
en->ru: 64.3 (train: genuine in-domain MCSQ data)
ru->en: 74.7 (train: additional backtranslated in-domain MCSQ data)
(Evaluated using multeval: https://github.com/jhclark/multeval)
MTMonkey is a web service which handles and distributes JSON-encoded HTTP requests for machine translation (MT) among multiple machines running an MT system, including text pre- and post processing.
It consists of an application server and remote workers which handle text processing and communicate translation requests to MT systems. The communication between the application server and the workers is based on the XML-RPC protocol. and The research leading to these results has received funding from the European Union Seventh Framework Programme (FP7/2007-2013) under grant agreement n° 257528 (KHRESMOI). This work has been using language resources developed and/or stored and/or distributed by the LINDAT-Clarin project of the Ministry of Education of the Czech Republic (project LM2010013). This work has been supported by the AMALACH grant (DF12P01OVV02) of the Ministry of Culture of the Czech Republic.
Data
-----
We have collected English-Odia parallel data for the purposes of NLP
research of the Odia language.
The data for the parallel corpus was extracted from existing parallel
corpora such as OdiEnCorp 1.0 and PMIndia, and books which contain both
English and Odia text such as grammar and bilingual literature books. We
also included parallel text from multiple public websites such as Odia
Wikipedia, Odia digital library, and Odisha Government websites.
The parallel corpus covers many domains: the Bible, other literature,
Wiki data relating to many topics, Government policies, and general
conversation. We have processed the raw data collected from the books,
websites, performed sentence alignments (a mix of manual and automatic
alignments) and released the corpus in a form suitable for various NLP
tasks.
Corpus Format
-------------
OdiEnCorp 2.0 is stored in simple tab-delimited plain text files, each
with three tab-delimited columns:
- a coarse indication of the domain
- the English sentence
- the corresponding Odia sentence
The corpus is shuffled at the level of sentence pairs.
The coarse domains are:
books ... prose text
dict ... dictionaries and phrasebooks
govt ... partially formal text
odiencorp10 ... OdiEnCorp 1.0 (mix of domains)
pmindia ... PMIndia (the original corpus)
wikipedia ... sentences and phrases from Wikipedia
Data Statistics
---------------
The statistics of the current release are given below.
Note that the statistics differ from those reported in the paper due to
deduplication at the level of sentence pairs. The deduplication was
performed within each of the dev set, test set and training set and
taking the coarse domain indication into account. It is still possible
that the same sentence pair appears more than once within the same set
(dev/test/train) if it came from different domains, and it is also
possible that a sentence pair appears in several sets (dev/test/train).
Parallel Corpus Statistics
--------------------------
Dev Dev Dev Test Test Test Train Train Train
Sents # EN # OD Sents # EN # OD Sents # EN # OD
books 3523 42011 36723 3895 52808 45383 3129 40461 35300
dict 3342 14580 13838 3437 14807 14110 5900 21591 20246
govt - - - - - - 761 15227 13132
odiencorp10 947 21905 19509 1259 28473 24350 26963 704114 602005
pmindia 3836 70282 61099 3836 68695 59876 30687 551657 486636
wikipedia 1896 9388 9385 1917 21381 20951 1930 7087 7122
Total 13544 158166 140554 14344 186164 164670 69370 1340137 1164441
"Sents" are the counts of the sentence pairs in the given set (dev/test/train)
and domain (books/dict/...).
"# EN" and "# OD" are approximate counts of words (simply space-delimited,
without tokenization) in English and Odia
The total number of sentence pairs (lines) is 13544+14344+69370=97258. Ignoring
the set and domain and deduplicating again, this number drops to 94857.
Citation
--------
If you use this corpus, please cite the following paper:
@inproceedings{parida2020odiencorp,
title={OdiEnCorp 2.0: Odia-English Parallel Corpus for Machine Translation},
author={Parida, Shantipriya and Dash, Satya Ranjan and Bojar, Ond{\v{r}}ej and Motlicek, Petr and Pattnaik, Priyanka and Mallick, Debasish Kumar},
booktitle={Proceedings of the WILDRE5--5th Workshop on Indian Language Data: Resources and Evaluation},
pages={14--19},
year={2020}
}
The January 2018 release of the ParaCrawl is the first version of the corpus. It contains parallel corpora for 11 languages paired with English, crawled from a large number of web sites. The selection of websites is based on CommonCrawl, but ParaCrawl is extracted from a brand new crawl which has much higher coverage of these selected websites than CommonCrawl. Since the data is fairly raw, it is released with two quality metrics that can be used for corpus filtering. An official "clean" version of each corpus uses one of the metrics. For more details and raw data download please visit: http://paracrawl.eu/releases.html
Statistical component of Chimera, a state-of-the-art MT system. and Project DF12P01OVV022 of the Ministry of Culture of the Czech Republic (NAKI -- Amalach).
The dataset used for the Ptakopět experiment on outbound machine translation. It consists of screenshots of web forms with user queries entered. The queries are available also in a text form. The dataset comprises two language versions: English and Czech. Whereas the English version has been fully post-processed (screenshots cropped, queries within the screenshots highlighted, dataset split based on its quality etc.), the Czech version is raw as it was collected by the annotators.
Post-editing and MQM annotations produced by the QT21 project. As described in
@InProceedings{specia-etal_MTSummit:2017,
author = {Specia, Lucia and Kim Harris and Frédéric Blain and Aljoscha Burchardt and Viviven Macketanz and Inguna Skadiņa and Matteo Negri and and Marco Turchi},
title = {Translation Quality and Productivity: A Study on Rich Morphology Languages},
booktitle = {Proceedings of Machine Translation Summit XVI},
year = {2017},
pages = {55--71},
address = {Nagoya, Japan},
}
This submission contains Dockerfile for creating a Docker image with compiled Tensor2tensor backend with compatible (TensorFlow Serving) models available in the Lindat Translation service (https://lindat.mff.cuni.cz/services/transformer/). Additionally, the submission contains a web frontend for simple in-browser access to the dockerized backend service.
Tensor2Tensor (https://github.com/tensorflow/tensor2tensor) is a library of deep learning models and datasets designed to make deep learning more accessible and accelerate ML research.
Test data for the WMT 2017 Automatic post-editing task (the same used for the Sentence-level Quality Estimation task). They consist in German-English triplets (source and target) belonging to the pharmacological domain and already tokenized. Test set contains 2,000 pairs. All data is provided by the EU project QT21 (http://www.qt21.eu/).
Test data for the WMT 2017 Automatic post-editing task (the same used for the Sentence-level Quality Estimation task). They consist in 2,000 English-German pairs (source and target) belonging to the IT domain and already tokenized. All data is provided by the EU project QT21 (http://www.qt21.eu/).
Test data for the WMT 2018 Automatic post-editing task. They consist in English-German pairs (source and target) belonging to the information technology domain and already tokenized. Test set contains 1,023 pairs. A neural machine translation system has been used to generate the target segments. All data is provided by the EU project QT21 (http://www.qt21.eu/).
Test data for the WMT 2018 Automatic post-editing task. They consist in English-German pairs (source and target) belonging to the information technology domain and already tokenized. Test set contains 2,000 pairs. A phrase-based machine translation system has been used to generate the target segments. This test set is sampled from the same dataset used for the 2016 and 2017 APE shared task editions. All data is provided by the EU project QT21 (http://www.qt21.eu/).
AMALACH project component TMODS:ENG-CZE; machine translation of queries from Czech to English. This archive contains models for the Moses decoder (binarized, pruned to allow for real-time translation) and configuration files for the MTMonkey toolkit. The aim of this package is to provide a full service for Czech->English translation which can be easily utilized as a component in a larger software solution. (The required tools are freely available and an installation guide is included in the package.)
The translation models were trained on CzEng 1.0 corpus and Europarl. Monolingual data for LM estimation additionally contains WMT news crawls until 2013.
En-De translation models, exported via TensorFlow Serving, available in the Lindat translation service (https://lindat.mff.cuni.cz/services/translation/).
Models are compatible with Tensor2tensor version 1.6.6.
For details about the model training (data, model hyper-parameters), please contact the archive maintainer.
Evaluation on newstest2020 (BLEU):
en->de: 25.9
de->en: 33.4
(Evaluated using multeval: https://github.com/jhclark/multeval)
En-Ru translation models, exported via TensorFlow Serving, available in the Lindat translation service (https://lindat.mff.cuni.cz/services/translation/).
Models are compatible with Tensor2tensor version 1.6.6.
For details about the model training (data, model hyper-parameters), please contact the archive maintainer.
Evaluation on newstest2020 (BLEU):
en->ru: 18.0
ru->en: 30.4
(Evaluated using multeval: https://github.com/jhclark/multeval)
This is the first release of the UFAL Parallel Corpus of North Levantine, compiled by the Institute of Formal and Applied Linguistics (ÚFAL) at Charles University within the Welcome project (https://welcome-h2020.eu/). The corpus consists of 120,600 multiparallel sentences in English, French, German, Greek, Spanish, and Standard Arabic selected from the OpenSubtitles2018 corpus [1] and manually translated into the North Levantine Arabic language. The corpus was created for the purpose of training machine translation for North Levantine and the other languages.
Data from a questionnaire survey conducted from 2022-08-25 to 2022-11-15 and exploring the use of machine translation by Ukrainian refugees in the Czech Republic. The presented spreadsheet contains minimally processed data exported from the two questionnaires that were created in Google Forms in the Ukrainian and the Russian language. The links to these questionnaires were distributed by three methods: direct email to particular refugees whose contact details the authors obtained while volunteering; through a non-profit organisation helping refugees (Vesna women’s education institution) and on social networks by posting links to the survey in groups associating the Ukrainian community across Czech regions and towns.
Since we asked potential respondents to spread the questionnaire further, we could not prevent it from reaching Ukrainians who had arrived in Czechia previously, or received temporary protection in other countries. Due to this fact, the textual answers to the question 1.5 "Which country are you in right now?" were replaced in the dataset by numbers (1 for Czech Republic, 2 for other countries) in order for us to be able to separate the data of respondents not located in the Czech Republic, which were irrelevant for our survey.
Training, development and text data (the same used for the Sentence-level Quality Estimation task) consist in English-German triplets (source, target and post-edit) belonging to the IT domain and already tokenized.
Training and development respectively contain 12,000 and 1,000 triplets, while the test set 2,000 instances. All data is provided by the EU project QT21 (http://www.qt21.eu/).
Training, development and test data consist in German sentences belonging to the IT domain and already tokenized. These sentences are the references of the data released for the 2016 edition of the WMT APE shared task. Differently from the data previously released, these sentences are obtained by manually translating the source sentence without leveraging the raw mt outputs. Training and development respectively contain 12,000 and 1,000 segments, while the test set 2,000 items. All data is provided by the EU project QT21 (http://www.qt21.eu/).
Training and development data for the WMT16 QE task. Test data will be published as a separate item.
This shared task will build on its previous four editions to further examine automatic methods for estimating the quality of machine translation output at run-time, without relying on reference translations. We include word-level, sentence-level and document-level estimation. The sentence and word-level tasks will explore a large dataset produced from post-editions by professional translators (as opposed to crowdsourced translations as in the previous year). For the first time, the data will be domain-specific (IT domain). The document-level task will use, for the first time, entire documents, which have been human annotated for quality indirectly in two ways: through reading comprehension tests and through a two-stage post-editing exercise. Our tasks have the following goals:
- To advance work on sentence and word-level quality estimation by providing domain-specific, larger and professionally annotated datasets.
- To study the utility of detailed information logged during post-editing (time, keystrokes, actual edits) for different levels of prediction.
- To analyse the effectiveness of different types of quality labels provided by humans for longer texts in document-level prediction.
This year's shared task provides new training and test datasets for all tasks, and allows participants to explore any additional data and resources deemed relevant. A in-house MT system was used to produce translations for the sentence and word-level tasks, and multiple MT systems were used to produce translations for the document-level task. Therefore, MT system-dependent information will be made available where possible.
The item contains models to tune for the WMT16 Tuning shared task for Czech-to-English.
CzEng 1.6pre (http://ufal.mff.cuni.cz/czeng/czeng16pre) corpus is used for the training of the translation models. The data is tokenized (using Moses tokenizer), lowercased and sentences longer than 60 words and shorter than 4 words are removed before training. Alignment is done using fast_align (https://github.com/clab/fast_align) and the standard Moses pipeline is used for training.
Two 5-gram language models are trained using KenLM: one only using the CzEng English data and the other is trained using all available English mono data for WMT except Common Crawl.
Also included are two lexicalized bidirectional reordering models, word based and hierarchical, with msd conditioned on both source and target of processed CzEng.
This item contains models to tune for the WMT16 Tuning shared task for English-to-Czech.
CzEng 1.6pre (http://ufal.mff.cuni.cz/czeng/czeng16pre) corpus is used for the training of the translation models. The data is tokenized (using Moses tokenizer), lowercased and sentences longer than 60 words and shorter than 4 words are removed before training. Alignment is done using fast_align (https://github.com/clab/fast_align) and the standard Moses pipeline is used for training.
Two 5-gram language models are trained using KenLM: one only using the CzEng Czech data and the other is trained using all available Czech mono data for WMT except Common Crawl.
Also included are two lexicalized bidirectional reordering models, word based and hierarchical, with msd conditioned on both source and target of processed CzEng.
Training and development data for the WMT 2017 Automatic post-editing task (the same used for the Sentence-level Quality Estimation task). They consist in German-English triplets (source, target and post-edit) belonging to the pharmacological domain and already tokenized. Training and development respectively contain 25,000 and 1,000 triplets. All data is provided by the EU project QT21 (http://www.qt21.eu/).
Training data for the WMT 2017 Automatic post-editing task (the same used for the Sentence-level Quality Estimation task). They consist in 11,000 English-German triplets (source, target and post-edit) belonging to the IT domain and already tokenized. All data is provided by the EU project QT21 (http://www.qt21.eu/).
Training and development data for the WMT17 QE task. Test data will be published as a separate item.
This shared task will build on its previous five editions to further examine automatic methods for estimating the quality of machine translation output at run-time, without relying on reference translations. We include word-level, phrase-level and sentence-level estimation. All tasks will make use of a large dataset produced from post-editions by professional translators. The data will be domain-specific (IT and Pharmaceutical domains) and substantially larger than in previous years. In addition to advancing the state of the art at all prediction levels, our goals include:
- To test the effectiveness of larger (domain-specific and professionally annotated) datasets. We will do so by increasing the size of one of last year's training sets.
- To study the effect of language direction and domain. We will do so by providing two datasets created in similar ways, but for different domains and language directions.
- To investigate the utility of detailed information logged during post-editing. We will do so by providing post-editing time, keystrokes, and actual edits.
This year's shared task provides new training and test datasets for all tasks, and allows participants to explore any additional data and resources deemed relevant. A in-house MT system was used to produce translations for all tasks. MT system-dependent information can be made available under request. The data is publicly available but since it has been provided by our industry partners it is subject to specific terms and conditions. However, these have no practical implications on the use of this data for research purposes.
Test data for the WMT17 QE task. Train data can be downloaded from http://hdl.handle.net/11372/LRT-1974
This shared task will build on its previous five editions to further examine automatic methods for estimating the quality of machine translation output at run-time, without relying on reference translations. We include word-level, phrase-level and sentence-level estimation. All tasks will make use of a large dataset produced from post-editions by professional translators. The data will be domain-specific (IT and Pharmaceutical domains) and substantially larger than in previous years. In addition to advancing the state of the art at all prediction levels, our goals include:
- To test the effectiveness of larger (domain-specific and professionally annotated) datasets. We will do so by increasing the size of one of last year's training sets.
- To study the effect of language direction and domain. We will do so by providing two datasets created in similar ways, but for different domains and language directions.
- To investigate the utility of detailed information logged during post-editing. We will do so by providing post-editing time, keystrokes, and actual edits.
This year's shared task provides new training and test datasets for all tasks, and allows participants to explore any additional data and resources deemed relevant. A in-house MT system was used to produce translations for all tasks. MT system-dependent information can be made available under request. The data is publicly available but since it has been provided by our industry partners it is subject to specific terms and conditions. However, these have no practical implications on the use of this data for research purposes.
Training and development data for the WMT 2018 Automatic post-editing task. They consist in English-German triplets (source, target and post-edit) belonging to the information technology domain and already tokenized. Training and development respectively contain 13,442 and 1,000 triplets. A neural machine translation system has been used to generate the target segments. All data is provided by the EU project QT21 (http://www.qt21.eu/).
Test data for the WMT18 QE task. Train data can be downloaded from http://hdl.handle.net/11372/LRT-2619.
This shared task will build on its previous six editions to further examine automatic methods for estimating the quality of machine translation output at run-time, without relying on reference translations. We include word-level, phrase-level and sentence-level estimation. All tasks make use of datasets produced from post-editions by professional translators. The datasets are domain-specific (IT and life sciences/pharma domains) and extend from those used previous years with more instances and more languages. One important addition is that this year we also include datasets with neural MT outputs. In addition to advancing the state of the art at all prediction levels, our specific goals are:
To study the performance of quality estimation approaches on the output of neural MT systems. We will do so by providing datasets for two language language pairs where the same source segments are translated by both a statistical phrase-based and a neural MT system.
To study the predictability of deleted words, i.e. words that are missing in the MT output. TO do so, for the first time we provide data annotated for such errors at training time.
To study the effectiveness of explicitly assigned labels for phrases. We will do so by providing a dataset where each phrase in the output of a phrase-based statistical MT system was annotated by human translators.
To study the effect of different language pairs. We will do so by providing datasets created in similar ways for four language language pairs.
To investigate the utility of detailed information logged during post-editing. We will do so by providing post-editing time, keystrokes, and actual edits.
Measure progress over years at all prediction levels. We will do so by using last year's test set for comparative experiments.
In-house statistical and neural MT systems were built to produce translations for all tasks. MT system-dependent information can be made available under request. The data is publicly available but since it has been provided by our industry partners it is subject to specific terms and conditions. However, these have no practical implications on the use of this data for research purposes. Participants are allowed to explore any additional data and resources deemed relevant.
Training and development data for the WMT18 QE task. Test data will be published as a separate item.
This shared task will build on its previous six editions to further examine automatic methods for estimating the quality of machine translation output at run-time, without relying on reference translations. We include word-level, phrase-level and sentence-level estimation. All tasks make use of datasets produced from post-editions by professional translators. The datasets are domain-specific (IT and life sciences/pharma domains) and extend from those used previous years with more instances and more languages. One important addition is that this year we also include datasets with neural MT outputs. In addition to advancing the state of the art at all prediction levels, our specific goals are:
To study the performance of quality estimation approaches on the output of neural MT systems. We will do so by providing datasets for two language language pairs where the same source segments are translated by both a statistical phrase-based and a neural MT system.
To study the predictability of deleted words, i.e. words that are missing in the MT output. TO do so, for the first time we provide data annotated for such errors at training time.
To study the effectiveness of explicitly assigned labels for phrases. We will do so by providing a dataset where each phrase in the output of a phrase-based statistical MT system was annotated by human translators.
To study the effect of different language pairs. We will do so by providing datasets created in similar ways for four language language pairs.
To investigate the utility of detailed information logged during post-editing. We will do so by providing post-editing time, keystrokes, and actual edits.
Measure progress over years at all prediction levels. We will do so by using last year's test set for comparative experiments.
In-house statistical and neural MT systems were built to produce translations for all tasks. MT system-dependent information can be made available under request. The data is publicly available but since it has been provided by our industry partners it is subject to specific terms and conditions. However, these have no practical implications on the use of this data for research purposes. Participants are allowed to explore any additional data and resources deemed relevant.
Marian NMT model for Catalan to Occitan translation. It is a multi-task model, producing also a phonemic transcription of the Catalan source. The model was submitted to WMT'21 Shared Task on Multilingual Low-Resource Translation for Indo-European Languages as a CUNI-Contrastive system for Catalan to Occitan.
Marian NMT model for Catalan to Occitan translation. Primary CUNI submission for WMT21 Multilingual
Low-Resource Translation for Indo-European Languages Shared Task.