ON-TRAC Consortium for End-to-End and Simultaneous Speech Translation Challenge Tasks at IWSLT 2020

TL;DR

This paper presents the ON-TRAC Consortium's end-to-end and simultaneous speech translation systems for IWSLT 2020, utilizing data augmentation, ensembling, and wait-k models to improve translation quality and latency control.

Contribution

The paper introduces novel data augmentation, ensembling techniques, and a latency control algorithm for cascade speech translation systems using Transformer-based wait-k models.

Findings

Effective data augmentation and ensembling improved offline translation accuracy.

The latency control algorithm achieved a good trade-off between latency and translation quality.

Transformer-based wait-k models performed well in simultaneous translation tasks.

Abstract

This paper describes the ON-TRAC Consortium translation systems developed for two challenge tracks featured in the Evaluation Campaign of IWSLT 2020, offline speech translation and simultaneous speech translation. ON-TRAC Consortium is composed of researchers from three French academic laboratories: LIA (Avignon Universit\'e), LIG (Universit\'e Grenoble Alpes), and LIUM (Le Mans Universit\'e). Attention-based encoder-decoder models, trained end-to-end, were used for our submissions to the offline speech translation track. Our contributions focused on data augmentation and ensembling of multiple models. In the simultaneous speech translation track, we build on Transformer-based wait-k models for the text-to-text subtask. For speech-to-text simultaneous translation, we attach a wait-k MT system to a hybrid ASR system. We propose an algorithm to control the latency of the ASR+MT cascade and achieve a good latency-quality trade-off on both subtasks.