UPDATE: Zenodo migration postponed to Oct 13 from 06:00-08:00 UTC. Read the announcement.

Conference paper Open Access

Samsung's System for the IWSLT 2019 End-to-End Speech Translation Task

Potapczyk, Tomasz; Przybysz, Pawel; Chochowski, Marcin; Szumaczuk, Artur

This paper describes the submission to IWSLT 2019 End- to-End speech translation task by Samsung R&D Institute, Poland. We decided to focus on end-to-end English to German TED lectures translation and did not provide any submission for other speech tasks. We used a slightly altered Transformer architecture with standard convolutional layer preparing the audio input to Transformer en- coder. Additionally, we propose an audio segmentation al- gorithm maximizing BLEU score on tst2015 test set.

Files (104.9 kB)
Name Size
IWSLT2019_paper_31.pdf
md5:236e612f3383a1ec298769550370d411
104.9 kB Download
565
429
views
downloads
All versions This version
Views 565565
Downloads 429429
Data volume 45.0 MB45.0 MB
Unique views 503503
Unique downloads 396396

Share

Cite as