@inproceedings{dey-etal-2022-clean,
    title = "Clean Text and Full-Body Transformer: {M}icrosoft`s Submission to the {WMT}22 Shared Task on Sign Language Translation",
    author = "Dey, Subhadeep  and
      Pal, Abhilash  and
      Chaabani, Cyrine  and
      Koller, Oscar",
    editor = {Koehn, Philipp  and
      Barrault, Lo{\"i}c  and
      Bojar, Ond{\v{r}}ej  and
      Bougares, Fethi  and
      Chatterjee, Rajen  and
      Costa-juss{\`a}, Marta R.  and
      Federmann, Christian  and
      Fishel, Mark  and
      Fraser, Alexander  and
      Freitag, Markus  and
      Graham, Yvette  and
      Grundkiewicz, Roman  and
      Guzman, Paco  and
      Haddow, Barry  and
      Huck, Matthias  and
      Jimeno Yepes, Antonio  and
      Kocmi, Tom  and
      Martins, Andr{\'e}  and
      Morishita, Makoto  and
      Monz, Christof  and
      Nagata, Masaaki  and
      Nakazawa, Toshiaki  and
      Negri, Matteo  and
      N{\'e}v{\'e}ol, Aur{\'e}lie  and
      Neves, Mariana  and
      Popel, Martin  and
      Turchi, Marco  and
      Zampieri, Marcos},
    booktitle = "Proceedings of the Seventh Conference on Machine Translation (WMT)",
    month = dec,
    year = "2022",
    address = "Abu Dhabi, United Arab Emirates (Hybrid)",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/jlcl-multiple-ingestion/2022.wmt-1.93/",
    pages = "969--976",
    abstract = "This paper describes Microsoft`s submission to the first shared task on sign language translation at WMT 2022, a public competition tackling sign language to spoken language translation for Swiss German sign language. The task is very challenging due to data scarcity and an unprecedented vocabulary size of more than 20k words on the target side. Moreover, the data is taken from real broadcast news, includes native signing and covers scenarios of long videos. Motivated by recent advances in action recognition, we incorporate full body information by extracting features from a pre-trained I3D model and applying a standard transformer network. The accuracy of the system is furtherimproved by applying careful data cleaning on the target text. We obtain BLEU scores of 0.6 and 0.78 on the test and dev set respectively, which is the best score among the participants of the shared task. Also in the human evaluation the submission reaches the first place. The BLEU score is further improved to 1.08 on the dev set by applying features extracted from a lip reading model."
}