ON-TRAC Consortium End-to-End Speech Translation Systems for the IWSLT 2019 Shared Task
This addresses speech translation for specific language pairs, but it is incremental as it builds on existing methods for shared task evaluation.
The paper tackled end-to-end speech translation for English-to-Portuguese by developing a neural encoder-decoder model, achieving BLEU scores of 26.91 on MuST-C and 43.82 on How2 validation sets.
This paper describes the ON-TRAC Consortium translation systems developed for the end-to-end model task of IWSLT Evaluation 2019 for the English-to-Portuguese language pair. ON-TRAC Consortium is composed of researchers from three French academic laboratories: LIA (Avignon Université), LIG (Université Grenoble Alpes), and LIUM (Le Mans Université). A single end-to-end model built as a neural encoder-decoder architecture with attention mechanism was used for two primary submissions corresponding to the two EN-PT evaluations sets: (1) TED (MuST-C) and (2) How2. In this paper, we notably investigate impact of pooling heterogeneous corpora for training, impact of target tokenization (characters or BPEs), impact of speech input segmentation and we also compare our best end-to-end model (BLEU of 26.91 on MuST-C and 43.82 on How2 validation sets) to a pipeline (ASR+MT) approach.