Worse WER, but Better BLEU? Leveraging Word Embedding as Intermediate in Multitask End-to-End Speech Translation
arXiv:2005.10678
Abstract
Speech translation (ST) aims to learn transformations from speech in the source language to the text in the target language. Previous works show that multitask learning improves the ST performance, in which the recognition decoder generates the text of the source language, and the translation decoder obtains the final translations based on the output of the recognition decoder. Because whether the output of the recognition decoder has the correct semantics is more critical than its accuracy, we propose to improve the multitask ST model by utilizing word embedding as the intermediate.
arXiv commentAccepted by ACL 2020
References in corpus (9)
- Enriching Word Vectors with Subword Information
- Listen, Attend and Spell
- Word Translation Without Parallel Data
- Unsupervised Machine Translation Using Monolingual Corpora Only
- Rethinking Feature Discrimination and Polymerization for Large-scale Recognition
- Learning Deep Features via Congenerous Cosine Loss for Person Recognition
- Von Mises-Fisher Loss for Training Sequence to Sequence Models with Continuous Outputs
- Sequence-to-Sequence Models Can Directly Translate Foreign Speech
- Multilingual End-to-End Speech Translation