Abstract
All of our systems start by transcribing the audio into text through an automatic speech recognition model trained on the TED-LIUM English Speech Recognition Corpus. Afterwards, we feed the transcripts into English-German text-based neural machine translation (NMT) models. Our systems employ three different translation models trained on separate training sets compiled from the English-German part of the TED Speech Translation Corpus and the OpenSubtitles2018 section of the OPUS collection.
In this paper, we also describe the experiments leading up to our final systems. Our experiments indicate that using OpenSubtitles2018 in training significantly improves translation performance. We also experimented with various pre- and postprocessing routines for the NMT module, but we did not have much success with these.
Our best-scoring system attains a BLEU score of 16.45 on the test set for this year’s task.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 15th International Workshop on Spoken Language Translation (IWSLT 2018) |
| Editors | Marco Turchi, Jan Niehues, Marcello Frederico |
| Number of pages | 6 |
| Place of Publication | Bruges |
| Publication date | 30 Oct 2018 |
| Pages | 89-94 |
| Publication status | Published - 30 Oct 2018 |
| MoE publication type | D3 Professional conference proceedings |
| Event | International Workshop on Spoken Language Translation - Brugge, Belgium Duration: 29 Oct 2018 → 30 Oct 2018 Conference number: 15 |
Fields of Science
- 113 Computer and information sciences
- 6121 Languages
Cite this
}
The MeMAD Submission to the IWSLT 2018 Speech Translation Task. / Sulubacak, Umut; Tiedemann, Jörg; Rouhe, Aku; Stig-Arne, Grönroos; Kurimo, Mikko.
Proceedings of the 15th International Workshop on Spoken Language Translation (IWSLT 2018). ed. / Marco Turchi; Jan Niehues; Marcello Frederico. Bruges, 2018. p. 89-94.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Professional
TY - GEN
T1 - The MeMAD Submission to the IWSLT 2018 Speech Translation Task
AU - Sulubacak, Umut
AU - Tiedemann, Jörg
AU - Rouhe, Aku
AU - Stig-Arne, Grönroos
AU - Kurimo, Mikko
PY - 2018/10/30
Y1 - 2018/10/30
N2 - This paper describes the MeMAD project entry to the IWSLT Speech Translation Shared Task, addressing the translation of English audio into German text. Between the pipeline and end-to-end model tracks, we participated only in the former, with three contrastive systems. We tried also the latter, but were not able to finish our end-to-end model in time.All of our systems start by transcribing the audio into text through an automatic speech recognition model trained on the TED-LIUM English Speech Recognition Corpus. Afterwards, we feed the transcripts into English-German text-based neural machine translation (NMT) models. Our systems employ three different translation models trained on separate training sets compiled from the English-German part of the TED Speech Translation Corpus and the OpenSubtitles2018 section of the OPUS collection.In this paper, we also describe the experiments leading up to our final systems. Our experiments indicate that using OpenSubtitles2018 in training significantly improves translation performance. We also experimented with various pre- and postprocessing routines for the NMT module, but we did not have much success with these.Our best-scoring system attains a BLEU score of 16.45 on the test set for this year’s task.
AB - This paper describes the MeMAD project entry to the IWSLT Speech Translation Shared Task, addressing the translation of English audio into German text. Between the pipeline and end-to-end model tracks, we participated only in the former, with three contrastive systems. We tried also the latter, but were not able to finish our end-to-end model in time.All of our systems start by transcribing the audio into text through an automatic speech recognition model trained on the TED-LIUM English Speech Recognition Corpus. Afterwards, we feed the transcripts into English-German text-based neural machine translation (NMT) models. Our systems employ three different translation models trained on separate training sets compiled from the English-German part of the TED Speech Translation Corpus and the OpenSubtitles2018 section of the OPUS collection.In this paper, we also describe the experiments leading up to our final systems. Our experiments indicate that using OpenSubtitles2018 in training significantly improves translation performance. We also experimented with various pre- and postprocessing routines for the NMT module, but we did not have much success with these.Our best-scoring system attains a BLEU score of 16.45 on the test set for this year’s task.
KW - 113 Computer and information sciences
KW - 6121 Languages
UR - https://workshop2018.iwslt.org/downloads/Proceedings_IWSLT_2018.pdf
M3 - Conference contribution
SP - 89
EP - 94
BT - Proceedings of the 15th International Workshop on Spoken Language Translation (IWSLT 2018)
A2 - Turchi, Marco
A2 - Niehues, Jan
A2 - Frederico, Marcello
CY - Bruges
ER -