Cross Attention Augmented Transducer

AbstractThis paper proposes a novel architecture, Cross Attention Augmented Transducer (CAAT), for simultaneous translation. The framework aims to jointly optimize the policy and translation models. To effectively consider all possible …...

AbstractThis paper describes USTC-NELSLIP’s submissions to the IWSLT2021 Simultaneous Speech Translation task. We proposed a novel simultaneous translation model, Cross-Attention Augmented Transducer (CAAT), which extends conventional RNN-T to sequence-to-sequence tasks …...